25:["$","title",null,{"children":"De la sospecha al tamizaje: perfiles asociados con HbA1c elevada | Leibnihtz Ayamamani-Choque | Observable"}] 49:Ta5d,{ const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px") .style("padding", "28px 0 10px") .style("line-height", "1.45"); container.append("div") .style("font-size", "13px") .style("font-weight", "700") .style("letter-spacing", "0.06em") .style("text-transform", "uppercase") .style("color", theme.blue) .text("Trabajo final | Big Data y Ciencia de Datos"); container.append("h1") .style("font-size", "42px") .style("line-height", "1.05") .style("max-width", "920px") .style("margin", "12px 0") .text("De la sospecha al tamizaje: perfiles asociados con HbA1c elevada"); container.append("p") .style("font-size", "19px") .style("max-width", "880px") .style("color", theme.slate) .text("La diabetes exige deteccion oportuna. Pero en campanas de salud o escenarios con recursos limitados, medir HbA1c a todos puede no ser inmediato. Este laboratorio explora si datos simples, y luego glucosa de campo, pueden ayudar a priorizar pruebas confirmatorias."); const cards = container.append("div") .style("display", "grid") .style("grid-template-columns", "repeat(3, minmax(0, 1fr))") .style("gap", "14px") .style("margin", "24px 0"); const data = [ {k: "Problema", v: "Deteccion temprana", d: "Priorizar quien deberia pasar a prueba confirmatoria."}, {k: "Datos", v: "NHANES 2017-2018", d: "Fuente abierta para construir un laboratorio metodologico."}, {k: "Salida", v: "Tres estados", d: "Baja, media/no concluyente y alta prioridad."} ]; const card = cards.selectAll("div") .data(data) .join("div") .style("border", `1px solid ${theme.border}`) .style("border-radius", "8px") .style("padding", "16px") .style("background", theme.bg); card.append("div") .style("font-size", "12px") .style("font-weight", "700") .style("color", theme.muted) .text(d => d.k); card.append("div") .style("font-size", "22px") .style("font-weight", "800") .style("margin", "6px 0") .text(d => d.v); card.append("div") .style("font-size", "13px") .style("color", theme.slate) .text(d => d.d); container.append("blockquote") .style("border-left", `5px solid ${theme.blue}`) .style("background", "#eff6ff") .style("padding", "14px 18px") .style("margin", "20px 0 0") .text("Pregunta analitica: En que medida las caracteristicas demograficas, corporales, de presion arterial y glucosa de campo ayudan a identificar perfiles con HbA1c elevada?"); return container.node(); }4a:T15be,{ const diabetesPeru = [ { territorio: "Costa", diagnostico_diabetes: 6.8, medicamentos: 72.7 }, { territorio: "Sierra", diagnostico_diabetes: 3.0, medicamentos: 53.4 }, { territorio: "Selva", diagnostico_diabetes: 4.1, medicamentos: 73.8 } ]; const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px") .style("line-height", "1.5"); container.append("h2").text("Diabetes y acceso a pruebas en el Peru"); container.append("p") .style("font-size", "16px") .style("color", theme.slate) .text("Segun la ENDES 2023:"); const cards = container.append("div") .style("display", "grid") .style("grid-template-columns", "repeat(3, minmax(0, 1fr))") .style("gap", "14px") .style("margin", "20px 0"); const cardData = [ { titulo: "Diagnostico declarado", valor: "5.5 %", detalle: "Poblacion peruana de 15 años o mas" }, { titulo: "Personas estimadas", valor: "1.3 M", detalle: "Aproximadamente, no 1.85 M" }, { titulo: "Acceso a medicamentos", valor: "70.3 %", detalle: "Diagnosticados que recibieron o compraron medicamentos" } ]; const card = cards.selectAll("div") .data(cardData) .join("div") .style("border", `1px solid ${theme.border}`) .style("border-radius", "8px") .style("padding", "16px") .style("background", theme.bg); card.append("div") .style("font-size", "12px") .style("font-weight", "700") .style("color", theme.muted) .text(d => d.titulo); card.append("div") .style("font-size", "30px") .style("font-weight", "800") .style("margin", "6px 0") .text(d => d.valor); card.append("div") .style("font-size", "13px") .style("color", theme.slate) .text(d => d.detalle); const width = 940; const height = 360; const margin = {top: 35, right: 30, bottom: 60, left: 70}; container.append("h3") .style("margin-top", "24px") .text("Diagnostico declarado y acceso a medicamentos por region natural"); const chartData = diabetesPeru.map(d => ({ territorio: d.territorio, diagnostico: d.diagnostico_diabetes, diagnostico_con_medicamentos: d.diagnostico_diabetes * d.medicamentos / 100, medicamentos: d.medicamentos })); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const x0 = d3.scaleBand() .domain(diabetesPeru.map(d => d.territorio)) .range([margin.left, width - margin.right]) .padding(0.35); const y = d3.scaleLinear() .domain([0, d3.max(chartData, d => d.diagnostico)]).nice() .range([height - margin.bottom, margin.top]); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x0)); svg.append("g") .attr("transform", `translate(${margin.left},0)`); svg.selectAll(".diagnostico-total") .data(chartData) .join("rect") .attr("class", "diagnostico-total") .attr("x", d => x0(d.territorio)) .attr("y", d => y(d.diagnostico)) .attr("width", x0.bandwidth()) .attr("height", d => y(0) - y(d.diagnostico)) .attr("fill", "#bfdbfe") .attr("stroke", theme.blue); svg.selectAll(".diagnostico-medicamentos") .data(chartData) .join("rect") .attr("class", "diagnostico-medicamentos") .attr("x", d => x0(d.territorio)) .attr("y", d => y(d.diagnostico_con_medicamentos)) .attr("width", x0.bandwidth()) .attr("height", d => y(0) - y(d.diagnostico_con_medicamentos)) .attr("fill", theme.green) .attr("opacity", 0.85); svg.selectAll(".label-total") .data(chartData) .join("text") .attr("class", "label-total") .attr("x", d => x0(d.territorio) + x0.bandwidth() / 2) .attr("y", d => y(d.diagnostico) - 8) .attr("text-anchor", "middle") .attr("font-size", 11) .text(d => `${d.diagnostico.toFixed(1)}% diag.`); svg.selectAll(".label-medicamentos") .data(chartData) .join("text") .attr("class", "label-medicamentos") .attr("x", d => x0(d.territorio) + x0.bandwidth() / 2) .attr("y", d => y(d.diagnostico_con_medicamentos) + 16) .attr("text-anchor", "middle") .attr("font-size", 11) .attr("fill", "white") .attr("font-weight", "700") .text(d => `${d.medicamentos.toFixed(1)}% med.`); const legend = container.append("div") .style("display", "flex") .style("gap", "18px") .style("margin", "8px 0 20px"); legend.selectAll("div") .data([ {label: "Total diagnosticado sobre poblacion 15+", color: "#bfdbfe"}, {label: "Subconjunto de diagnosticados que recibio/compro medicamentos", color: theme.green} ]) .join("div") .style("display", "flex") .style("align-items", "center") .style("gap", "6px") .html(d => `${d.label}`); container.append("blockquote") .style("border-left", `5px solid ${theme.blue}`) .style("background", "#eff6ff") .style("padding", "12px 16px") .style("margin", "18px 0") .text("Puente metodologico: NHANES no representa al Peru, pero permite construir y demostrar un pipeline reproducible para priorizar evaluacion confirmatoria cuando medir HbA1c a todos no es inmediato."); return container.node(); }4b:Tdce,{ const data = trazabilidad.map(d => ({ archivo: d.archivo, registros_originales: +d.registros_originales, columnas_originales: +d.columnas_originales, registros_despues_seleccion: +d.registros_despues_seleccion, registros_con_match_demo: +d.registros_con_match_demo, registros_base_demo_despues_union_left: +d.registros_base_demo_despues_union_left, registros_dataset_analitico: +d.registros_dataset_analitico, registros_analiticos_con_variable_clave: +d.registros_analiticos_con_variable_clave, porcentaje_coincidencia_seqn: +d.porcentaje_coincidencia_seqn })); const seleccionado = data.find(d => d.archivo === archivoSeleccionado); const width = 940; const height = 330; const margin = {top: 28, right: 20, bottom: 70, left: 80}; const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px"); container.append("h2").text("Pipeline de datos: de componentes NHANES a registro analitico"); container.append("p") .style("color", theme.slate) .text("Antes del modelo, verificamos que las uniones sean trazables y que cada componente aporte informacion al registro analitico."); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const x = d3.scaleBand() .domain(data.map(d => d.archivo)) .range([margin.left, width - margin.right]) .padding(0.25); const y = d3.scaleLinear() .domain([0, d3.max(data, d => d.registros_originales)]).nice() .range([height - margin.bottom, margin.top]); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x)) .selectAll("text") .attr("transform", "rotate(-30)") .style("text-anchor", "end"); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(y).ticks(6)); svg.selectAll("rect") .data(data) .join("rect") .attr("x", d => x(d.archivo)) .attr("y", d => y(d.registros_originales)) .attr("width", x.bandwidth()) .attr("height", d => y(0) - y(d.registros_originales)) .attr("fill", d => d.archivo === archivoSeleccionado ? theme.blue : "#bfdbfe"); svg.selectAll(".label") .data(data) .join("text") .attr("x", d => x(d.archivo) + x.bandwidth() / 2) .attr("y", d => y(d.registros_originales) - 6) .attr("text-anchor", "middle") .attr("font-size", 11) .text(d => formatNum(d.registros_originales)); const detail = container.append("div") .style("display", "grid") .style("grid-template-columns", "repeat(4, minmax(0, 1fr))") .style("gap", "12px") .style("margin-top", "18px"); const detailData = [ ["Registros originales", formatNum(seleccionado.registros_originales)], ["Columnas originales", seleccionado.columnas_originales], ["Match con DEMO", formatNum(seleccionado.registros_con_match_demo)], ["Coincidencia SEQN", `${seleccionado.porcentaje_coincidencia_seqn.toFixed(1)}%`] ]; const card = detail.selectAll("div") .data(detailData) .join("div") .style("border", `1px solid ${theme.border}`) .style("border-radius", "8px") .style("padding", "12px") .style("background", theme.bg); card.append("div").style("font-size", "12px").style("color", theme.muted).text(d => d[0]); card.append("div").style("font-size", "22px").style("font-weight", "800").text(d => d[1]); return container.node(); }4c:T7f0,{ const data = edaDistribuciones .filter(d => d.variable === variableDistribucion) .map(d => ({ categoria: d.rango_o_categoria, porcentaje: +d.porcentaje, frecuencia: +d.frecuencia })); const width = 940; const height = 360; const margin = {top: 30, right: 30, bottom: 90, left: 80}; const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px"); container.append("h2").text("Perfil de la muestra analitica"); container.append("p") .style("color", theme.slate) .text("La prevalencia mostrada es descriptiva de la muestra analitica. No debe confundirse automaticamente con una estimacion poblacional."); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const x = d3.scaleBand() .domain(data.map(d => d.categoria)) .range([margin.left, width - margin.right]) .padding(0.2); const y = d3.scaleLinear() .domain([0, d3.max(data, d => d.porcentaje)]).nice() .range([height - margin.bottom, margin.top]); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x)) .selectAll("text") .attr("transform", "rotate(-30)") .style("text-anchor", "end"); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(y).tickFormat(d => `${d}%`)); svg.selectAll("rect") .data(data) .join("rect") .attr("x", d => x(d.categoria)) .attr("y", d => y(d.porcentaje)) .attr("width", x.bandwidth()) .attr("height", d => y(0) - y(d.porcentaje)) .attr("fill", theme.blue); svg.selectAll(".label") .data(data) .join("text") .attr("x", d => x(d.categoria) + x.bandwidth() / 2) .attr("y", d => y(d.porcentaje) - 6) .attr("text-anchor", "middle") .attr("font-size", 11) .text(d => `${d.porcentaje.toFixed(1)}%`); return container.node(); }4d:Te67,{ const xMap = { imc: "imc", cintura: "cintura_cm", glucosa: "glucosa", edad: "edad" }; const xVar = xMap[ejeXScatter]; const data = edaScatter .map(d => ({ x: d[xVar] == null || d[xVar] === "" ? NaN : Number(d[xVar]), y: d.hba1c == null || d.hba1c === "" ? NaN : Number(d.hba1c), clase: d.clase, sexo: d.sexo })) .filter(d => Number.isFinite(d.x) && Number.isFinite(d.y) && d.x > 0) const width = 940; const height = 440; const margin = {top: 30, right: 30, bottom: 60, left: 70}; const container = d3.create("section") .style("position", "relative") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px"); container.append("h2").text("EDA: las clases se superponen"); container.append("p") .style("color", theme.slate) .text("Vemos participantes con perfiles similares pero distinto estado de HbA1c, por eso una decision binaria estricta es limitada."); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const x = d3.scaleLinear() .domain(d3.extent(data, d => d.x)).nice() .range([margin.left, width - margin.right]); const y = d3.scaleLinear() .domain(d3.extent(data, d => d.y)).nice() .range([height - margin.bottom, margin.top]); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x)); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(y)); svg.append("line") .attr("x1", margin.left) .attr("x2", width - margin.right) .attr("y1", y(5.7)) .attr("y2", y(5.7)) .attr("stroke", theme.red) .attr("stroke-dasharray", "5,5"); svg.append("text") .attr("x", width - margin.right) .attr("y", y(5.7) - 8) .attr("text-anchor", "end") .attr("font-size", 12) .attr("fill", theme.red) .text("Umbral HbA1c = 5.7"); svg.selectAll("circle") .data(data) .join("circle") .attr("cx", d => x(d.x)) .attr("cy", d => y(d.y)) .attr("r", 3.5) .attr("fill", d => d.clase === "HbA1c elevada" ? theme.red : theme.blue) .attr("opacity", 0.6) .on("mouseenter", function(event, d) { d3.select(this) .attr("r", 6) .attr("opacity", 1) .attr("stroke", "#111827") .attr("stroke-width", 1.5); tooltip .style("opacity", 1) .html(` ${d.clase}
${ejeXScatter}: ${d.x.toFixed(1)}
HbA1c: ${d.y.toFixed(1)} %
Sexo: ${d.sexo ?? "Sin dato"} `); }) .on("mousemove", function(event) { tooltip .style("left", `${event.offsetX + 18}px`) .style("top", `${event.offsetY + 18}px`); }) .on("mouseleave", function() { d3.select(this) .attr("r", 3.5) .attr("opacity", 0.6) .attr("stroke", null); tooltip.style("opacity", 0); }); svg.append("text") .attr("x", width / 2) .attr("y", height - 12) .attr("text-anchor", "middle") .attr("font-size", 13) .text(ejeXScatter); svg.append("text") .attr("transform", "rotate(-90)") .attr("x", -height / 2) .attr("y", 18) .attr("text-anchor", "middle") .attr("font-size", 13) .text("HbA1c (%)"); const tooltip = container.append("div") .style("position", "absolute") .style("pointer-events", "none") .style("background", "rgba(15, 23, 42, 0.92)") .style("color", "white") .style("padding", "8px 10px") .style("border-radius", "6px") .style("font-size", "12px") .style("line-height", "1.35") .style("opacity", 0); return container.node(); }4e:T9f9,{ const width = 520; const height = 430; const margin = {top: 50, right: 30, bottom: 70, left: 90}; const container = d3.create("section") .style("font-family", "system-ui, sans-serif") .style("max-width", "620px"); container.append("h2") .text(`Matriz de confusión: ${modeloMatriz}`); container.append("p") .style("color", theme.slate) .text("La matriz muestra aciertos y errores del modelo en el conjunto de prueba. La clase 1 corresponde a HbA1c elevada."); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const clases = ["0", "1"]; const x = d3.scaleBand() .domain(clases) .range([margin.left, width - margin.right]) .padding(0.08); const y = d3.scaleBand() .domain(clases) .range([margin.top, height - margin.bottom]) .padding(0.08); const color = d3.scaleSequential() .domain([0, d3.max(matrizModelo, d => d.cantidad)]) .interpolator(d3.interpolateBlues); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x).tickFormat(d => d === "1" ? "Predicha elevada" : "Predicha no elevada")); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(y).tickFormat(d => d === "1" ? "Real elevada" : "Real no elevada")); svg.selectAll("rect") .data(matrizModelo) .join("rect") .attr("x", d => x(d.clase_predicha)) .attr("y", d => y(d.clase_real)) .attr("width", x.bandwidth()) .attr("height", y.bandwidth()) .attr("fill", d => color(d.cantidad)) .attr("stroke", "#fff"); svg.selectAll("text.cell") .data(matrizModelo) .join("text") .attr("class", "cell") .attr("x", d => x(d.clase_predicha) + x.bandwidth() / 2) .attr("y", d => y(d.clase_real) + y.bandwidth() / 2) .attr("text-anchor", "middle") .attr("dominant-baseline", "middle") .attr("font-size", 18) .attr("font-weight", "700") .attr("fill", d => d.cantidad > d3.max(matrizModelo, x => x.cantidad) * 0.55 ? "white" : "#111827") .text(d => d.cantidad); svg.append("text") .attr("x", width / 2) .attr("y", height - 18) .attr("text-anchor", "middle") .attr("font-weight", "700") .text("Clase predicha"); svg.append("text") .attr("transform", "rotate(-90)") .attr("x", -height / 2) .attr("y", 24) .attr("text-anchor", "middle") .attr("font-weight", "700") .text("Clase real"); return container.node(); }4f:Ta17,{ const data = comparacionGlucosa .filter(d => ["f1", "roc_auc", "recall", "precision"].includes(d.metrica)) .flatMap(d => [ {metrica: d.metrica, escenario: "Base sin glucosa", valor: +d.modelo_base_sin_glucosa}, {metrica: d.metrica, escenario: "Campo con glucosa", valor: +d.modelo_campo_con_glucosa} ]); const width = 940; const height = 390; const margin = {top: 40, right: 30, bottom: 80, left: 80}; const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px"); container.append("h2").text("La glucosa ayuda, pero no resuelve todo"); container.append("p") .style("color", theme.slate) .text("El glucometro mejora F1, ROC-AUC y precision. Sin embargo, la reduccion de falsos negativos es pequena, por eso no reemplaza HbA1c confirmatoria."); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const x0 = d3.scaleBand() .domain([...new Set(data.map(d => d.metrica))]) .range([margin.left, width - margin.right]) .padding(0.22); const x1 = d3.scaleBand() .domain(["Base sin glucosa", "Campo con glucosa"]) .range([0, x0.bandwidth()]) .padding(0.08); const y = d3.scaleLinear() .domain([0, d3.max(data, d => d.valor)]).nice() .range([height - margin.bottom, margin.top]); const color = d3.scaleOrdinal() .domain(["Base sin glucosa", "Campo con glucosa"]) .range(["#93c5fd", theme.green]); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x0)); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(y)); svg.append("g") .selectAll("g") .data(d3.group(data, d => d.metrica)) .join("g") .attr("transform", ([metrica]) => `translate(${x0(metrica)},0)`) .selectAll("rect") .data(([, values]) => values) .join("rect") .attr("x", d => x1(d.escenario)) .attr("y", d => y(d.valor)) .attr("width", x1.bandwidth()) .attr("height", d => y(0) - y(d.valor)) .attr("fill", d => color(d.escenario)); svg.append("g") .selectAll("text") .data(data) .join("text") .attr("x", d => x0(d.metrica) + x1(d.escenario) + x1.bandwidth() / 2) .attr("y", d => y(d.valor) - 6) .attr("text-anchor", "middle") .attr("font-size", 11) .text(d => d.metrica === "falsos_negativos" ? d.valor.toFixed(0) : d.valor.toFixed(3)); return container.node(); }50:T95b,{ const data = resumenTamizaje .filter(d => d.escenario === escenarioTamizaje) .map(d => ({ banda: d.banda_probabilidad, n: +d.n, porcentaje: +d.porcentaje_muestra, positivos: +d.positivos_reales, prevalencia: +d.prevalencia_hba1c_elevada })); const width = 940; const height = 360; const margin = {top: 30, right: 30, bottom: 85, left: 80}; const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px"); container.append("p") .style("font-size", "13px") .style("color", theme.muted) .text("Nota metodológica: esta visualización usa únicamente el conjunto de prueba/test (25 %) de la submuestra con HbA1c y glucosa disponibles. Las barras suman 608 participantes."); container.append("h2").text("Salida de tamizaje: baja, media y alta prioridad"); container.append("p") .style("color", theme.slate) .text("Distribución de participantes del conjunto test en tres zonas de decisión. La etiqueta muestra cantidad de personas y porcentaje real con HbA1c elevada dentro de cada zona."); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const x = d3.scaleBand() .domain(data.map(d => d.banda)) .range([margin.left, width - margin.right]) .padding(0.25); const y = d3.scaleLinear() .domain([0, d3.max(data, d => d.n)]).nice() .range([height - margin.bottom, margin.top]); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x)) .selectAll("text") .attr("transform", "rotate(-25)") .style("text-anchor", "end"); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(y)); svg.selectAll("rect") .data(data) .join("rect") .attr("x", d => x(d.banda)) .attr("y", d => y(d.n)) .attr("width", x.bandwidth()) .attr("height", d => y(0) - y(d.n)) .attr("fill", d => colorEstado(d.banda)); svg.selectAll(".label") .data(data) .join("text") .attr("x", d => x(d.banda) + x.bandwidth() / 2) .attr("y", d => y(d.n) - 8) .attr("text-anchor", "middle") .attr("font-size", 12) .text(d => `${d.n} | ${d.prevalencia.toFixed(1)}% elev.`); return container.node(); }51:T852,{ const data = curvaRocEscenarios.map(d => ({ escenario: d.escenario, fpr: +d.fpr, tpr: +d.tpr })); const width = 680; const height = 480; const margin = {top: 30, right: 30, bottom: 55, left: 65}; const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "760px"); container.append("h2").text("ROC-AUC: capacidad de ordenar riesgo"); container.append("p") .style("color", theme.slate) .text("La glucosa mejora la discriminacion global, aunque no elimina falsos negativos."); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const x = d3.scaleLinear().domain([0, 1]).range([margin.left, width - margin.right]); const y = d3.scaleLinear().domain([0, 1]).range([height - margin.bottom, margin.top]); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x)); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(y)); svg.append("line") .attr("x1", x(0)) .attr("x2", x(1)) .attr("y1", y(0)) .attr("y2", y(1)) .attr("stroke", "#94a3b8") .attr("stroke-dasharray", "4,4"); const line = d3.line() .x(d => x(d.fpr)) .y(d => y(d.tpr)); const color = d3.scaleOrdinal() .domain(["modelo_base_sin_glucosa", "modelo_campo_con_glucosa"]) .range([theme.blue, theme.green]); svg.append("g") .selectAll("path") .data(d3.group(data, d => d.escenario)) .join("path") .attr("fill", "none") .attr("stroke", ([escenario]) => color(escenario)) .attr("stroke-width", 3) .attr("d", ([, values]) => line(values)); svg.append("text") .attr("x", width / 2) .attr("y", height - 12) .attr("text-anchor", "middle") .text("False positive rate"); svg.append("text") .attr("transform", "rotate(-90)") .attr("x", -height / 2) .attr("y", 18) .attr("text-anchor", "middle") .text("True positive rate"); return container.node(); }52:T7e4,{ const data = importanciaEscenarios .filter(d => d.escenario === escenarioImportancia) .sort((a, b) => +a.ranking - +b.ranking) .slice(0, 10) .map(d => ({ variable: d.variable.replace("numeric__", "").replace("categorical__", ""), odds: +d.odds_ratio, coef: +d.coeficiente })) .reverse(); const width = 940; const height = 420; const margin = {top: 30, right: 40, bottom: 50, left: 260}; const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px"); container.append("h2").text("Interpretabilidad: variables asociadas, no causas"); container.append("p") .style("color", theme.slate) .text("Los odds ratios ayudan a interpretar direccion e intensidad de asociacion predictiva. No prueban causalidad ni reemplazan criterio clinico."); const svg = container.append("svg") .attr("viewBox", [0, 0, width, height]) .style("width", "100%") .style("height", "auto"); const x = d3.scaleLinear() .domain([0, d3.max(data, d => d.odds)]).nice() .range([margin.left, width - margin.right]); const y = d3.scaleBand() .domain(data.map(d => d.variable)) .range([height - margin.bottom, margin.top]) .padding(0.2); svg.append("g") .attr("transform", `translate(0,${height - margin.bottom})`) .call(d3.axisBottom(x)); svg.append("g") .attr("transform", `translate(${margin.left},0)`) .call(d3.axisLeft(y)); svg.append("line") .attr("x1", x(1)) .attr("x2", x(1)) .attr("y1", margin.top) .attr("y2", height - margin.bottom) .attr("stroke", "#64748b") .attr("stroke-dasharray", "4,4"); svg.selectAll("rect") .data(data) .join("rect") .attr("x", x(0)) .attr("y", d => y(d.variable)) .attr("width", d => x(d.odds) - x(0)) .attr("height", y.bandwidth()) .attr("fill", d => d.coef >= 0 ? theme.red : theme.blue) .attr("opacity", 0.8); return container.node(); }53:T597,{ const container = d3.create("section") .style("font-family", "system-ui, -apple-system, Segoe UI, sans-serif") .style("max-width", "1050px") .style("line-height", "1.5"); container.append("h2").text("Conclusion: una herramienta de priorizacion, no de diagnostico"); const bullets = [ "El modelo base muestra que variables simples contienen senal, pero no separan perfectamente las clases.", "Agregar glucosa mejora F1 y ROC-AUC, especialmente la precision, pero reduce poco los falsos negativos.", "Por eso la salida de tres estados es mas responsable que una clasificacion binaria rigida.", "La zona media/no concluyente debe comunicarse como decision pendiente, no como error.", "La recomendacion final es priorizar HbA1c confirmatoria en alta y media probabilidad, y mantener seguimiento preventivo en baja probabilidad." ]; const list = container.append("ul") .style("font-size", "17px"); list.selectAll("li") .data(bullets) .join("li") .style("margin", "10px 0") .text(d => d); container.append("blockquote") .style("border-left", `5px solid ${theme.red}`) .style("background", "#fef2f2") .style("padding", "14px 18px") .style("margin-top", "22px") .text("Demostracion academica basada en datos poblacionales. No constituye diagnostico medico, recomendacion clinica ni evaluacion individual de salud."); return container.node(); }54:[{"id":"9d6982f89e89f47efdfa43417c6248f768bfe7de82c27135507cf2c0988dfb6ad7c689f4941377ab5db893166165ab93a0580435941006b23111f767551b3a93","href":"https://static.observableusercontent.com/files/9d6982f89e89f47efdfa43417c6248f768bfe7de82c27135507cf2c0988dfb6ad7c689f4941377ab5db893166165ab93a0580435941006b23111f767551b3a93","name":"diabetes-simulador-hba1c.html","type":"text/html","lastModified":1784145321966.015,"size":17409},{"id":"102d4a76b8ff8d0170bfd684697c0121feb3b8e4fa82574ec8a68e5221d21a3294bfa98180de66fc1a14256347dbfb0aea2b74288800a26369b927fcca4427a0","href":"https://static.observableusercontent.com/files/102d4a76b8ff8d0170bfd684697c0121feb3b8e4fa82574ec8a68e5221d21a3294bfa98180de66fc1a14256347dbfb0aea2b74288800a26369b927fcca4427a0","name":"calidad_dama_mapa.csv","type":"text/csv","lastModified":1784140179227.968,"size":3502},{"id":"ec353c47eac125ad9bca9cdc2a0d5563461f97696d3ebc275a2343bf8b9f745328d7aab6757265a52af1af61f9cbc759e65f7404bdc0a3ad540468d99a09fbdb","href":"https://static.observableusercontent.com/files/ec353c47eac125ad9bca9cdc2a0d5563461f97696d3ebc275a2343bf8b9f745328d7aab6757265a52af1af61f9cbc759e65f7404bdc0a3ad540468d99a09fbdb","name":"eda_segmentos.csv","type":"text/csv","lastModified":1784140229233.076,"size":3405},{"id":"7f12ce051b24da9c4d63d3f83dc15260fa167c52bc78653ab5a803d10f6f2734176070f58c91937bcad8dc4ee3b2dac7bb063ac5a3aae2f50e4f16fef20635b0","href":"https://static.observableusercontent.com/files/7f12ce051b24da9c4d63d3f83dc15260fa167c52bc78653ab5a803d10f6f2734176070f58c91937bcad8dc4ee3b2dac7bb063ac5a3aae2f50e4f16fef20635b0","name":"calidad_variables.csv","type":"text/csv","lastModified":1784161758474.782,"size":3823},{"id":"b5f02c9c07ecad65daf095f247445300010a75daaa802ba21860070beb0500b6f945d13384d2f76867a04322608858e1213e982ca98ea4adb1482b5bbfb365fa","href":"https://static.observableusercontent.com/files/b5f02c9c07ecad65daf095f247445300010a75daaa802ba21860070beb0500b6f945d13384d2f76867a04322608858e1213e982ca98ea4adb1482b5bbfb365fa","name":"matriz_confusion.csv","type":"text/csv","lastModified":1784163622217.995,"size":550},{"id":"3afe4f264fb69f36904c4b2724ae710b62fc53b0d04d5225e3d3e8b4e644ac335f1f57cc9cda64a9b33a7c00ec40d4aefd7b9ca9624559278f3d1d518c2327c1","href":"https://static.observableusercontent.com/files/3afe4f264fb69f36904c4b2724ae710b62fc53b0d04d5225e3d3e8b4e644ac335f1f57cc9cda64a9b33a7c00ec40d4aefd7b9ca9624559278f3d1d518c2327c1","name":"modelo_nhanes_web.onnx","type":"application/octet-stream","lastModified":1784163991464.582,"size":765},{"id":"a7f0e26715636cbb2e9c28ac103c02cd32f82c3ca09b7d373e14c61ac602715494770d5b177b85f6b39274b7969b1f9354b980cdeafcade25a7be1a7afe4ec42","href":"https://static.observableusercontent.com/files/a7f0e26715636cbb2e9c28ac103c02cd32f82c3ca09b7d373e14c61ac602715494770d5b177b85f6b39274b7969b1f9354b980cdeafcade25a7be1a7afe4ec42","name":"metadata_proyecto.csv","type":"text/csv","lastModified":1784140157692.007,"size":502},{"id":"0de31ba23bd97ca171daa083706757d348dd1f7737f2d7f74802f6cea108cddf6668a7f4cb3bcb50a2020988cf3bfdb8f221e64eafadcacf7aef36963f07bac5","href":"https://static.observableusercontent.com/files/0de31ba23bd97ca171daa083706757d348dd1f7737f2d7f74802f6cea108cddf6668a7f4cb3bcb50a2020988cf3bfdb8f221e64eafadcacf7aef36963f07bac5","name":"eda_distribuciones.csv","type":"text/csv","lastModified":1784140208896.191,"size":1113},{"id":"d4a4234fe16a1abc03ef48e5ae6f2846b8f8d94207fd58f88b693b73b3d4c84d5cfb12265d2130ed213ce69f3d8ef30ad9d3d4248ce761e51579b7a6ef8f6c55","href":"https://static.observableusercontent.com/files/d4a4234fe16a1abc03ef48e5ae6f2846b8f8d94207fd58f88b693b73b3d4c84d5cfb12265d2130ed213ce69f3d8ef30ad9d3d4248ce761e51579b7a6ef8f6c55","name":"eda_scatter_muestra.csv","type":"text/csv","lastModified":1784140246534.404,"size":62000},{"id":"d94e941bbc236c308af793d870e279721abe4284400dca3b77231d846a6baeba940e001474f1a7c29e335ff5c1da24bc09378f1866b86e042a6d8e94d54b6408","href":"https://static.observableusercontent.com/files/d94e941bbc236c308af793d870e279721abe4284400dca3b77231d846a6baeba940e001474f1a7c29e335ff5c1da24bc09378f1866b86e042a6d8e94d54b6408","name":"costo_operativo_tamizaje.csv","type":"text/csv","lastModified":1784140334393.165,"size":713},{"id":"260d6fc7928f3c22b5e757e66505f5be2260a7a0ee5afdb51a9d49b85c89c8bd9ec361908de6a9383742295cb780a7e201c739a9729684149a05a47c3406d7ea","href":"https://static.observableusercontent.com/files/260d6fc7928f3c22b5e757e66505f5be2260a7a0ee5afdb51a9d49b85c89c8bd9ec361908de6a9383742295cb780a7e201c739a9729684149a05a47c3406d7ea","name":"comparacion_modelo_base_vs_glucosa.csv","type":"text/csv","lastModified":1784140263763.935,"size":549},{"id":"36f05503fd9f5e353ff219b9def8e49e40bbccbe9340f7cac86d4db6efd84aaa929df46c1e5351a0baa18abf372d8765d0aa49f95b30e438888e5dd8265ce8de","href":"https://static.observableusercontent.com/files/36f05503fd9f5e353ff219b9def8e49e40bbccbe9340f7cac86d4db6efd84aaa929df46c1e5351a0baa18abf372d8765d0aa49f95b30e438888e5dd8265ce8de","name":"metricas_modelo_campo_glucosa.csv","type":"text/csv","lastModified":1784140282290.057,"size":533},{"id":"4172157aa7d84044ad612403bb3c5072de7a080f9b241454f491b207ab58b0159b6a94515d6fcf7850b046ff8a5385b4253c20752f7aef5a3d34d060566eac95","href":"https://static.observableusercontent.com/files/4172157aa7d84044ad612403bb3c5072de7a080f9b241454f491b207ab58b0159b6a94515d6fcf7850b046ff8a5385b4253c20752f7aef5a3d34d060566eac95","name":"predicciones_tamizaje_tres_estados.csv","type":"text/csv","lastModified":1784140302057.563,"size":310957},{"id":"581f4b8660f1deaa2be8956c23c822e4cbdb3accb79d1ec0cda94ca374e01e1c0f31078b8e18eea960ddf7386da527dc7fd6aca60174dae79e2bf74023c73033","href":"https://static.observableusercontent.com/files/581f4b8660f1deaa2be8956c23c822e4cbdb3accb79d1ec0cda94ca374e01e1c0f31078b8e18eea960ddf7386da527dc7fd6aca60174dae79e2bf74023c73033","name":"resumen_tamizaje_tres_estados.csv","type":"text/csv","lastModified":1784140314305.094,"size":868},{"id":"49d3171ec54459bf5123029d7eedd1726635a2d522335c53eeeafe9e4e70c218fe4127163c305d5f16c00f496604dbf2988ba6c0688cddc0ecdf30da33c18e46","href":"https://static.observableusercontent.com/files/49d3171ec54459bf5123029d7eedd1726635a2d522335c53eeeafe9e4e70c218fe4127163c305d5f16c00f496604dbf2988ba6c0688cddc0ecdf30da33c18e46","name":"riesgo_interpretacion_clinica.csv","type":"text/csv","lastModified":1784140348489.297,"size":801},{"id":"03238e530762abcb7266a32a045322c8916b4420f336ea1729169cc3ba8baa13a50d676c4f1c2d1ea30e9046fd54f863783ea82e1912db3380ef323b676cb8e4","href":"https://static.observableusercontent.com/files/03238e530762abcb7266a32a045322c8916b4420f336ea1729169cc3ba8baa13a50d676c4f1c2d1ea30e9046fd54f863783ea82e1912db3380ef323b676cb8e4","name":"importancia_variables_escenarios_glucosa.csv","type":"text/csv","lastModified":1784140363616.77,"size":1594},{"id":"956a981fda59c9c84283e2eb1d3a7aaa42dd3bd3039309d6875e030526d970f7e5e32667ef7cca6810c9a48bb47b36e21593965457a4a95a4101bf889667fba5","href":"https://static.observableusercontent.com/files/956a981fda59c9c84283e2eb1d3a7aaa42dd3bd3039309d6875e030526d970f7e5e32667ef7cca6810c9a48bb47b36e21593965457a4a95a4101bf889667fba5","name":"curva_roc_escenarios_glucosa.csv","type":"text/csv","lastModified":1784140383716.057,"size":22168},{"id":"ca097db1b5d044d9d6ecb551c1c09ff8f3ce39667aad4264e07c66cadd956c5cef3e260dabb23d95d87f28513025a10028dd663faaba00ec44bb602d6f416986","href":"https://static.observableusercontent.com/files/ca097db1b5d044d9d6ecb551c1c09ff8f3ce39667aad4264e07c66cadd956c5cef3e260dabb23d95d87f28513025a10028dd663faaba00ec44bb602d6f416986","name":"trazabilidad_pipeline@1.csv","type":"text/csv","lastModified":1784160792766.712,"size":743},{"id":"e491f819a53d2d3aa1a86b828f8909987f414632e702f3e9648d4af3c3495a8aedb4649af3caeef34709d1d38c2b1491f810c5108ff54ea49e3b95b361024966","href":"https://static.observableusercontent.com/files/e491f819a53d2d3aa1a86b828f8909987f414632e702f3e9648d4af3c3495a8aedb4649af3caeef34709d1d38c2b1491f810c5108ff54ea49e3b95b361024966","name":"diccionario_datos.csv","type":"text/csv","lastModified":1784161393893.762,"size":2353}] 29:{"id":"688e588d6ec24a16","slug":null,"title":"De la sospecha al tamizaje: perfiles asociados con HbA1c elevada","thumbnail":"373fd03ee6d6c05c1a2c7c03b708d2d9bc5c3d95dd2dc111e4a0dec9adc789bf","access_level":"public","create_time":"$D2026-07-15T18:28:42.159Z","update_time":"$D2026-07-16T01:16:20.539Z","view_time":null,"first_public_time":"$D2026-07-16T01:21:23.929Z","has_unpublished_edits":null,"fork":false,"stars":0,"starred":false,"owner":{"type":"team","id":"642f642bcad5a3a9","login":"leibnihtz","name":"Leibnihtz Ayamamani-Choque","avatar_url":"https://avatars-next.observableusercontent.com/125816f99fb99b0465bc7e35145c7ddc337013fb60d918338c76b9c6bc40254d","tier":"starter_2024","bio":"","followed_by_you":false,"follows_you":false},"version":350,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"70eae1d3f6d99dd3","login":"leibnihtz","name":"Leibnihtz Ayamamani-Choque","avatar_url":"https://avatars-next.observableusercontent.com/125816f99fb99b0465bc7e35145c7ddc337013fb60d918338c76b9c6bc40254d","tier":null,"bio":"","followed_by_you":false,"follows_you":false}],"members":[{"type":"team","id":"642f642bcad5a3a9","login":"leibnihtz","name":"Leibnihtz Ayamamani-Choque","avatar_url":"https://avatars-next.observableusercontent.com/125816f99fb99b0465bc7e35145c7ddc337013fb60d918338c76b9c6bc40254d","tier":"starter_2024","role":"editor"},{"type":"individual","id":"70eae1d3f6d99dd3","login":"leibnihtz","name":"Leibnihtz Ayamamani-Choque","avatar_url":"https://avatars-next.observableusercontent.com/125816f99fb99b0465bc7e35145c7ddc337013fb60d918338c76b9c6bc40254d","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"titling":"auto","title":"De la sospecha al tamizaje: perfiles asociados con HbA1c elevada","cells":[{"id":65,"value":"$49","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":155,"value":"`","pinned":false,"hidden":false,"mode":"html","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":69,"value":"$4a","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":75,"value":"viewof archivoSeleccionado = Inputs.select(\n trazabilidad.map(d => d.archivo),\n {label: \"Archivo origen\", value: trazabilidad[0].archivo}\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":78,"value":"$4b","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":81,"value":"Inputs.table(trazabilidad, {\n columns: [\n \"archivo\",\n \"registros_originales\",\n \"registros_con_match_demo\",\n \"registros_dataset_analitico\",\n \"registros_analiticos_con_variable_clave\",\n ]\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":174,"value":"viewof rolAnalitico = Inputs.select(\n [\"Todos\", ...Array.from(new Set(diccionario.map(d => d.rol_analitico)))],\n {\n label: \"Rol analítico\",\n value: \"Todos\"\n }\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":183,"value":"Inputs.table(diccionarioFiltrado, {\n columns: [\n \"variable_original\",\n \"variable_final\",\n \"archivo_origen\",\n \"descripcion\",\n \"rol_analitico\"\n ]\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":190,"value":"viewof dimensionCalidad = Inputs.select(\n [\"Todas\", ...Array.from(new Set(calidadVariables.map(d => d.dimension)))],\n {\n label: \"Dimensión de calidad\",\n value: \"Todas\"\n }\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":196,"value":"md`### Calidad de datos: ${dimensionCalidad}\n\nSe muestran **${calidadVariablesFiltradas.length}** indicadores por variable.`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":199,"value":"Inputs.table(calidadVariablesFiltradas, {\n columns: [\n \"variable\",\n \"dimension\",\n \"indicador\",\n \"porcentaje\",\n \"umbral_esperado\",\n \"estado\"\n ]\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":207,"value":"Inputs.table(dimensionesDamaComplementarias, {\n columns: [\n \"dimension\",\n \"evidencia\",\n \"archivo\"\n ]\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":217,"value":"\n# EDA","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":96,"value":"viewof variableDistribucion = Inputs.select(\n Array.from(new Set(edaDistribuciones.map(d => d.variable))),\n {label: \"Variable\", value: \"clase_hba1c\"}\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":99,"value":"$4c","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":102,"value":"viewof ejeXScatter = Inputs.select(\n [\"imc\", \"cintura\", \"glucosa\", \"edad\"],\n {label: \"Variable X\", value: \"glucosa\"}\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":105,"value":"$4d","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":283,"value":"matrizConfusion = FileAttachment(\"matriz_confusion.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":286,"value":"viewof modeloMatriz = Inputs.select(\n Array.from(new Set(matrizConfusion.map(d => d.modelo))),\n {\n label: \"Modelo\",\n value: \"regresion_logistica\"\n }\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":289,"value":"matrizModelo = matrizConfusion\n .filter(d => d.modelo === modeloMatriz)\n .map(d => ({\n clase_real: String(d.clase_real),\n clase_predicha: String(d.clase_predicha),\n cantidad: +d.cantidad,\n porcentaje: +d.porcentaje\n }))","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":293,"value":"$4e","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":342,"value":"## Prueba interactiva del modelo ONNX \n\nEl modelo exportado a formato **ONNX**, es un estándar que permite ejecutar modelos de Machine Learning fuera de Python.\n\nCon esos datos estima la probabilidad de pertenecer al grupo con **HbA1c elevada**.\n\nEsta prueba no busca diagnosticar diabetes. Su propósito es demostrar que el pipeline completo puede pasar de:\n\n**Datos originales -> Preparación -> Entrenamiento -> Exportación -> Simulación interactiva**","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":232,"value":"viewof edadSim = Inputs.range([20, 80], {\n label: \"Edad\",\n value: 50,\n step: 1\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":235,"value":"viewof sexoSim = Inputs.select([\"Hombre\", \"Mujer\"], {\n label: \"Sexo\",\n value: \"Hombre\"\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":238,"value":"viewof imcSim = Inputs.range([18, 45], {\n label: \"IMC\",\n value: 28,\n step: 0.1\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":241,"value":"viewof cinturaSim = Inputs.range([65, 140], {\n label: \"Cintura (cm)\",\n value: 95,\n step: 0.5\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":244,"value":"viewof tallaSim = Inputs.range([140, 200], {\n label: \"Talla de referencia (cm)\",\n value: 170,\n step: 1\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":247,"value":"viewof sistolicaSim = Inputs.range([90, 180], {\n label: \"Presión sistólica\",\n value: 125,\n step: 1\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":250,"value":"viewof diastolicaSim = Inputs.range([55, 110], {\n label: \"Presión diastólica\",\n value: 78,\n step: 1\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":253,"value":"viewof ingresoSim = Inputs.range([0, 5], {\n label: \"Ratio ingreso-pobreza\",\n value: 2.5,\n step: 0.05\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":301,"value":"entradaOnnx = Float32Array.from([\n edadSim,\n imcSim,\n cinturaSim,\n cinturaSim / tallaSim,\n sistolicaSim,\n diastolicaSim,\n ingresoSim,\n sexoSim === \"Hombre\" ? 1 : 2\n])","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":319,"value":"md`Probabilidad estimada: **${(probabilidadOnnx * 100).toFixed(1)}%**\n\nClase ONNX: **${claseOnnx}**`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":108,"value":"$4f","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":111,"value":"Inputs.table(comparacionGlucosa, {\n columns: [\n \"metrica\",\n \"modelo_base_sin_glucosa\",\n \"modelo_campo_con_glucosa\",\n \"diferencia_campo_menos_base\",\n \"interpretacion\"\n ]\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":114,"value":"viewof escenarioTamizaje = Inputs.select(\n Array.from(new Set(resumenTamizaje.map(d => d.escenario))),\n {label: \"Escenario\", value: \"modelo_campo_con_glucosa\", format: nombreEscenario}\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":117,"value":"$50","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":120,"value":"Inputs.table(resumenTamizaje.filter(d => d.escenario === escenarioTamizaje), {\n columns: [\n \"banda_probabilidad\",\n \"n\",\n \"porcentaje_muestra\",\n \"positivos_reales\",\n \"prevalencia_hba1c_elevada\",\n \"falsos_negativos_si_baja_no_se_prioriza\",\n \"mensaje_operativo\"\n ]\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":130,"value":"Inputs.table(riesgoClinico, {\n columns: [\n \"escenario\",\n \"riesgo_principal\",\n \"nivel_riesgo_interpretacion\",\n \"motivo\",\n \"f1\",\n \"roc_auc\",\n \"falsos_negativos\",\n \"mitigacion\"\n ]\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":133,"value":"$51","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":136,"value":"viewof escenarioImportancia = Inputs.select(\n Array.from(new Set(importanciaEscenarios.map(d => d.escenario))),\n {label: \"Escenario\", value: \"modelo_campo_con_glucosa\", format: nombreEscenario}\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":139,"value":"$52","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":142,"value":"$53","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":229,"value":"ort = import(\"https://esm.sh/onnxruntime-web@1.18.0\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":296,"value":"ortWeb = new Promise((resolve, reject) => {\n const script = document.createElement(\"script\");\n script.src = \"https://cdn.jsdelivr.net/npm/onnxruntime-web@1.18.0/dist/ort.min.js\";\n script.onload = () => resolve(window.ort);\n script.onerror = reject;\n document.head.appendChild(script);\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":323,"value":"sessionOnnx = await ortWeb.InferenceSession.create(\n await FileAttachment(\"modelo_nhanes_web.onnx\").arrayBuffer(),\n { executionProviders: [\"wasm\"] }\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":327,"value":"tensorOnnx = new ortWeb.Tensor(\"float32\", entradaOnnx, [1, 8])","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":331,"value":"resultadoOnnx = await sessionOnnx.run({\n input: tensorOnnx\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":335,"value":"probabilidadOnnx = resultadoOnnx.probabilities.data[1]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":339,"value":"claseOnnx = resultadoOnnx.label.data[0]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":0,"value":"d3 = require(\"d3@7\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":8,"value":"metadata = FileAttachment(\"metadata_proyecto.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":11,"value":"calidadDamaMapa = FileAttachment(\"calidad_dama_mapa.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":14,"value":"trazabilidad = FileAttachment(\"trazabilidad_pipeline@1.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":17,"value":"edaDistribuciones = FileAttachment(\"eda_distribuciones.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":20,"value":"edaSegmentos = FileAttachment(\"eda_segmentos.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":23,"value":"edaScatter = FileAttachment(\"eda_scatter_muestra.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":26,"value":"comparacionGlucosa = FileAttachment(\"comparacion_modelo_base_vs_glucosa.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":29,"value":"metricasCampo = FileAttachment(\"metricas_modelo_campo_glucosa.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":32,"value":"prediccionesTamizaje = FileAttachment(\"predicciones_tamizaje_tres_estados.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":35,"value":"resumenTamizaje = FileAttachment(\"resumen_tamizaje_tres_estados.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":38,"value":"costoTamizaje = FileAttachment(\"costo_operativo_tamizaje.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":41,"value":"riesgoClinico = FileAttachment(\"riesgo_interpretacion_clinica.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":44,"value":"importanciaEscenarios = FileAttachment(\"importancia_variables_escenarios_glucosa.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":47,"value":"curvaRocEscenarios = FileAttachment(\"curva_roc_escenarios_glucosa.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":50,"value":"theme = ({\n blue: \"#2563eb\",\n green: \"#16a34a\",\n amber: \"#d97706\",\n red: \"#dc2626\",\n slate: \"#334155\",\n muted: \"#64748b\",\n border: \"#e2e8f0\",\n bg: \"#f8fafc\"\n})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":53,"value":"formatPct = d3.format(\".1%\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":56,"value":"formatNum = d3.format(\",\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":59,"value":"colorEstado = estado =>\n estado === \"alta_probabilidad_priorizar_hba1c\" ? theme.red :\n estado === \"probabilidad_media_no_concluyente\" ? theme.amber :\n estado === \"baja_probabilidad\" ? theme.green :\n theme.blue","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":62,"value":"nombreEscenario = escenario =>\n escenario === \"modelo_base_sin_glucosa\"\n ? \"Modelo base sin glucosa\"\n : \"Modelo de campo con glucosa\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":172,"value":"diccionario = FileAttachment(\"diccionario_datos.csv\").csv({typed: true})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":178,"value":"diccionarioFiltrado = rolAnalitico === \"Todos\"\n ? diccionario\n : diccionario.filter(d => d.rol_analitico === rolAnalitico)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":188,"value":"calidadVariables = FileAttachment(\"calidad_variables.csv\").csv({typed: true})","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":192,"value":"calidadVariablesFiltradas = dimensionCalidad === \"Todas\"\n ? calidadVariables\n : calidadVariables.filter(d => d.dimension === dimensionCalidad)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":204,"value":"dimensionesDamaComplementarias = [\n {\n dimension: \"Unicidad\",\n evidencia: \"1 row por participante\",\n archivo: \"trazabilidad_pipeline.csv\"\n },\n {\n dimension: \"Integridad\",\n evidencia: \"Se validó la disponibilidad de variables clave.\",\n archivo: \"trazabilidad_pipeline.csv\"\n },\n {\n dimension: \"Oportunidad\",\n evidencia: \"Ciclo analizado corresponde a NHANES 2017-2018\",\n archivo: \"metadata_proyecto.csv\"\n }\n]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W54","stdlib":"1"}} 24:null