3c:null 3d:["$","title",null,{"children":"Pregnancy according to My Apple Watch | Sangyu Xu | Observable"}] 43:T5c2,md`# Pregnancy according to My Apple Watch In 2023, my mother and my partner conspired to gift me an Apple Watch for my birthday. I was skeptical: my mental model of a smartwatch was mostly a notification buzzer that needs charging more often than a phone. But I strapped it on, and within a few days the neuroscientist in me was genuinely absorbed by the data it was quietly accumulating. An accelerometer on my wrist recording my resting heart rate and estimating VO2 max through gait analysis isn't that different. The hardware is sleeker, and the subject is less amenable to being anaesthetized. At the time I was about three years past my first child, in reasonably good shape, and curious about what the watch would tell me. What I didn't anticipate was that it would witness my second pregnancy almost from the beginning through nine months of gestation, through birth and the postpartum period, and into a training recovery I'm still in the middle of. I also didn't anticipate that I would not sit down to look at all of this data in one place until considerably later than you'd expect from someone whose job involves looking at longitudinal data in one place. My analytical self emerged rather slowly from the postpartum haze. When I finally did dig up all the metrics, aligned on the same time axis, from 2023 through mid-2026, I was quite surprised by how clearly pregnant I was in the data. The watch just kept recording. Let's look at what it captured.` 44:T412,```js html`` ```45:T40b,function smoothSeries(rows, window = 30) { const observed = rows.filter((d) => Number.isFinite(d.value)); const byDay = new Map(observed.map((d) => [dayKey(d.date), d.value])); const start = d3.utcDay.floor(observed[0].date); const end = d3.utcDay.floor(observed.at(-1).date); const days = d3.utcDay.range(start, d3.utcDay.offset(end, 1)); const observedDays = observed.map((d) => d3.utcDay.floor(d.date)); let left = 0; const daily = days.map((date) => { const key = dayKey(date); if (byDay.has(key)) return {date, value: byDay.get(key)}; while (left < observedDays.length - 2 && observedDays[left + 1] < date) left += 1; const d0 = observedDays[left]; const d1 = observedDays[left + 1]; const v0 = observed[left].value; const v1 = observed[left + 1].value; const t = (date - d0) / (d1 - d0); return {date, value: v0 + (v1 - v0) * t}; }); const smoothed = hammingSmooth(daily.map((d) => d.value), window); return daily.map((d, i) => ({date: d.date, value: smoothed[i]})); }46:T79d,function plotMetric(metric, unit, width) { const raw = series[metric]; const smooth = smoothSeries(raw, 30); const xDomain = d3.extent(raw, (d) => d.date); const yDomain = d3.extent(raw.concat(smooth), (d) => d.value); const yPad = (yDomain[1] - yDomain[0]) * 0.08 || 1; const ymin = yDomain[0] - yPad; const ymax = yDomain[1] + yPad; const yLabel = ymax - (ymax - ymin) * 0.05; const eventLabel = ymin + (ymax - ymin) * 0.06; const spans = phaseSpans .map((span) => { const start = d3.max([span.start, xDomain[0]]); const end = d3.min([span.end, xDomain[1]]); return { ...span, start, end, mid: new Date((start.getTime() + end.getTime()) / 2), y1: ymin, y2: ymax, y: yLabel }; }) .filter((span) => span.start < span.end); const events = keyEvents .filter((event) => event.date >= xDomain[0] && event.date <= xDomain[1]) .map((event) => ({...event, y: eventLabel})); return Plot.plot({ width, height: 220, marginTop: 12, marginRight: 16, marginBottom: 44, marginLeft: 58, x: { domain: xDomain, ticks: d3.utcMonth.every(3), tickFormat: d3.utcFormat("%Y-%m"), label: null }, y: { domain: [ymin, ymax], grid: true, label: unit }, marks: [ Plot.rect(spans, {x1: "start", x2: "end", y1: "y1", y2: "y2", fill: "color", fillOpacity: 0.07}), Plot.text(spans, {x: "mid", y: "y", text: "short", fill: "color", fontSize: 9, fontWeight: 700}), Plot.ruleX(events, {x: "date", stroke: "#cfcfcf", strokeDasharray: "2,3"}), Plot.text(events, {x: "date", y: "y", text: "label", fill: "#999", fontSize: 8, rotate: -90, textAnchor: "end"}), Plot.dot(raw, {x: "date", y: "value", r: 1.55, fill: "#378ADD", fillOpacity: 0.14}), Plot.line(smooth, {x: "date", y: "value", stroke: "#184E8E", strokeWidth: 2.5}) ] }); }47:T1050,interactive3d = { function renderSideBySide(width) { const halfW = Math.max(320, Math.floor((width - 24) / 2)); // Shared render function (same as before but takes data + config directly) function makeTraces(rows, zColumn, zLabel, events) { return [ { type: "scatter3d", mode: "lines", name: "Trajectory", showlegend: false, x: rows.map(d => d.VO2Max), y: rows.map(d => d.RestingHR), z: rows.map(d => d[zColumn]), line: {color: "#b8b8b8", width: 3}, opacity: 0.35, hoverinfo: "skip" }, ...phaseOrder.map(phase => { const pr = rows.filter(d => d.phase === phase); return { type: "scatter3d", mode: "markers", name: phase, x: pr.map(d => d.VO2Max), y: pr.map(d => d.RestingHR), z: pr.map(d => d[zColumn]), text: pr.map(d => `${d.dateLabel}
${phase}`), marker: {size: 3.2, color: colors.get(phase), opacity: 0.78}, hovertemplate: `%{text}
VO2Max: %{x:.2f}
Resting HR: %{y:.1f}
${zLabel}: %{z:.4f}` }; }).filter(t => t.x.length), ...events.map(event => { const row = nearestRow(rows, event.date); return { type: "scatter3d", mode: "markers+text", showlegend: false, x: [row.VO2Max], y: [row.RestingHR], z: [row[zColumn]], text: [event.label], textposition: "top center", textfont: {color: event.color, size: 11}, marker: {color: event.color, size: 8, symbol: "diamond", line: {color: "white", width: 1.5}}, hoverinfo: "skip" }; }), // Gap line: Now → Peak { type: "scatter3d", mode: "lines", name: "Gap to peak", showlegend: false, x: [nearestRow(rows, parseDate("2026-07-01")).VO2Max, nearestRow(rows, parseDate("2023-06-14")).VO2Max], y: [nearestRow(rows, parseDate("2026-07-01")).RestingHR, nearestRow(rows, parseDate("2023-06-14")).RestingHR], z: [nearestRow(rows, parseDate("2026-07-01"))[zColumn], nearestRow(rows, parseDate("2023-06-14"))[zColumn]], line: {color: "#4954D6", width: 3, dash: "dash"}, opacity: 0.85, hoverinfo: "skip" } ]; } function makeLayout(title, zLabel, w) { return { title: {text: title, font: {size: 13, color: "#333"}}, width: w, height: Math.min(680, Math.max(380, Math.round(w * 0.9))), margin: {l: 40, r: 40, t: 44, b: 40}, paper_bgcolor: "white", legend: { x: 1.2, y: 1, bgcolor: "rgba(255,255,255,0.82)", borderwidth: 0, font: {size: 10}, itemsizing: "constant", itemwidth: 30 }, scene: { xaxis: {title: "VO2Max", gridcolor: "#e6e6e6", zerolinecolor: "#e6e6e6"}, yaxis: {title: "Resting HR", gridcolor: "#e6e6e6", zerolinecolor: "#e6e6e6"}, zaxis: {title: zLabel, gridcolor: "#e6e6e6", zerolinecolor: "#e6e6e6"}, camera: {eye: {"x":0.97,"y":-2.639,"z":1.65}} } }; } const config = {responsive: false, displaylogo: false}; const wrapper = document.createElement("div"); wrapper.style.cssText = "display:flex;gap:24px;flex-wrap:wrap;"; // Left: Walk Asymmetry const leftDiv = document.createElement("div"); leftDiv.style.cssText = `flex:1;min-width:320px;`; wrapper.appendChild(leftDiv); // Right: Ds SDNN const rightDiv = document.createElement("div"); rightDiv.style.cssText = `flex:1;min-width:320px;`; wrapper.appendChild(rightDiv); Plotly.newPlot( leftDiv, makeTraces(ssAsym, "WalkAsym", "Walk Asymmetry", key3dAsym), makeLayout("Walk Asymmetry", "Walk Asymmetry", halfW), config ); Plotly.newPlot( rightDiv, makeTraces(ssDbl, "DoubleSup", "Double Support", key3dDbl), makeLayout("Double Support Time", "Double Support", halfW), config ); return wrapper; } return renderSideBySide(width); } 48:T19ae,state_space_asymmetry = { const rows = ssAsym; const zColumn = "WalkAsym"; const zLabel = "Walk Asymmetry"; // Sample every 7 days for smooth but performant animation const allDates = rows.map(d => d.dateLabel); const frameDates = allDates.filter((_, i) => i % 7 === 0); // ── Traces ─────────────────────────────────────────────────────────────── // 1. Full grey trajectory (static) const trajectoryTrace = { type: "scatter3d", mode: "lines", name: "Full trajectory", showlegend: false, x: rows.map(d => d.VO2Max), y: rows.map(d => d.RestingHR), z: rows.map(d => d[zColumn]), line: {color: "#d8d8d8", width: 3}, opacity: 0.45, hoverinfo: "skip" }; // 2. Phase background dots (faded, for legend + context) const phaseTraces = phaseOrder.map(phase => { const pr = rows.filter(d => d.phase === phase); return { type: "scatter3d", mode: "markers", name: phase, x: pr.map(d => d.VO2Max), y: pr.map(d => d.RestingHR), z: pr.map(d => d[zColumn]), marker: {size: 3.2, color: colors.get(phase), opacity: 0.25}, hoverinfo: "skip" }; }).filter(t => t.x.length); // 3. "Me" dot — starts at frame 0 position const firstRow = rows.find(r => r.dateLabel === frameDates[0]) || rows[0]; const youTrace = { type: "scatter3d", mode: "markers", name: "Me", showlegend: true, x: [firstRow.VO2Max], y: [firstRow.RestingHR], z: [firstRow[zColumn]], marker: { color: colors.get(firstRow.phase), size: 12, symbol: "circle", line: {color: "gray", width: 3}, opacity: 1 }, hovertemplate: "You
VO2Max: %{x:.1f}
HR: %{y:.0f}
" + `${zLabel}: %{z:.5f}` } const allTraces = [trajectoryTrace, ...phaseTraces, youTrace]; const youIdx = allTraces.length - 1; // ── Layout ──────────────────────────────────────────────────────────────── const layout = { paper_bgcolor: "white", margin: {l: 5, r: 5, t: 12, b: 5}, legend: { x: 1, y: 0, bgcolor: "rgba(255,255,255,0.85)", borderwidth: 0, font: {size: 10}, itemsizing: "constant", itemwidth: 30 }, scene: { xaxis: {title: "VO2Max", gridcolor: "#e6e6e6", zerolinecolor: "#e6e6e6"}, yaxis: {title: "Resting HR", gridcolor: "#e6e6e6", zerolinecolor: "#e6e6e6"}, zaxis: {title: "Walk Asymmetry", gridcolor: "#e6e6e6", zerolinecolor: "#e6e6e6"}, camera: {eye: {"x":0.97,"y":-2.639,"z":1.65}} } }; // ── Container ───────────────────────────────────────────────────────────── const container = document.createElement("div"); // Controls row const controls = document.createElement("div"); controls.style.cssText = "display:flex;align-items:center;gap:12px;padding:10px 2px 6px;font-size:13px;font-family:sans-serif;"; const playBtn = document.createElement("button"); playBtn.textContent = "▶ Play"; playBtn.style.cssText = "padding:6px 16px;border-radius:7px;border:1.5px solid #ccc;" + "background:#fff;cursor:pointer;font-size:13px;font-weight:600;"; const resetBtn = document.createElement("button"); resetBtn.textContent = "↺ Reset"; resetBtn.style.cssText = "padding:6px 12px;border-radius:7px;border:1.5px solid #ccc;" + "background:#fff;cursor:pointer;font-size:13px;"; const dateSpan = document.createElement("span"); dateSpan.style.cssText = "color:#666;font-family:monospace;min-width:90px;"; dateSpan.textContent = frameDates[0]; const phaseSpan = document.createElement("span"); phaseSpan.style.cssText = "font-weight:700;min-width:140px;"; phaseSpan.textContent = firstRow.phase; phaseSpan.style.color = colors.get(firstRow.phase); controls.append(playBtn, resetBtn, dateSpan, phaseSpan); container.appendChild(controls); // Plot div const plotDiv = document.createElement("div"); plotDiv.style.cssText = `width:100%;height:${Math.min(700, Math.round(width * 0.72))}px;`; container.appendChild(plotDiv); await Plotly.newPlot(plotDiv, allTraces, layout, { responsive: true, displaylogo: false }); let cameraFrame = 0; const cameraInterval = setInterval(() => { const progress = currentFrame / (frameDates.length - 1); const theta = 2 * Math.PI * progress; Plotly.relayout(plotDiv, { "scene.camera.eye": { x: 2 * Math.cos(theta), y: 2 * Math.sin(theta), z: 1 } }); cameraFrame++; }, 50); // ── Animation logic ─────────────────────────────────────────────────────── let playing = false; let intervalId = null; let currentFrame = 0; function goToFrame(i) { const dateStr = frameDates[i]; const row = rows.find(r => r.dateLabel === dateStr) || rows[rows.length - 1]; if (!row) return; dateSpan.textContent = dateStr; phaseSpan.textContent = row.phase; phaseSpan.style.color = colors.get(row.phase); // Update dot position AND colour to match current phase Plotly.restyle(plotDiv, { x: [[row.VO2Max]], y: [[row.RestingHR]], z: [[row[zColumn]]], "marker.color": [colors.get(row.phase)] }, [youIdx]); } goToFrame(0); startAnimation(); function startAnimation() { if (playing) return; if (currentFrame >= frameDates.length - 1) currentFrame = 0; playing = true; playBtn.textContent = "⏸ Pause"; intervalId = setInterval(() => { currentFrame++; if (currentFrame >= frameDates.length) { playing = false; clearInterval(intervalId); playBtn.textContent = "▶ Play"; currentFrame = frameDates.length - 1; return; } goToFrame(currentFrame); }, 90); } playBtn.addEventListener("click", () => { if (playing) { playing = false; clearInterval(intervalId); playBtn.textContent = "▶ Play"; } else { startAnimation(); } }); resetBtn.addEventListener("click", () => { playing = false; clearInterval(intervalId); playBtn.textContent = "▶ Play"; currentFrame = 0; goToFrame(0); }); return container; } 49:[{"id":"645bda7a09ec1ccc183b084ae521eb80509a6ac3b684b41595b9b1f2aa6f1af44c596468347a04b98fd13d697daaca2fb16ccd9292f835a635c909b592758141","href":"https://static.observableusercontent.com/files/645bda7a09ec1ccc183b084ae521eb80509a6ac3b684b41595b9b1f2aa6f1af44c596468347a04b98fd13d697daaca2fb16ccd9292f835a635c909b592758141","name":"WalkingDoubleSupportPercentage_daily.csv","type":"text/csv","lastModified":1783248831496.271,"size":31560},{"id":"99e2cec727477db88f49e6d131fe1ecfa7cc3fe0232e79d6e5296132712c1982a8054af6a25c6c2bbe7e7dfbfa9e1548e1655b04a52f4538be747dd86021344a","href":"https://static.observableusercontent.com/files/99e2cec727477db88f49e6d131fe1ecfa7cc3fe0232e79d6e5296132712c1982a8054af6a25c6c2bbe7e7dfbfa9e1548e1655b04a52f4538be747dd86021344a","name":"health_metric_coverage.csv","type":"text/csv","lastModified":1783243115490.919,"size":522},{"id":"0c873e92c6383417f75c6e217c2ded0c0697c359f02ac7d6f6a7d87c4594295b6ba177e8f8fbf667dbdf9ef4d4922aa7af50e4e0b84099862a7e4a303ed85f1d","href":"https://static.observableusercontent.com/files/0c873e92c6383417f75c6e217c2ded0c0697c359f02ac7d6f6a7d87c4594295b6ba177e8f8fbf667dbdf9ef4d4922aa7af50e4e0b84099862a7e4a303ed85f1d","name":"HeartRateVariabilitySDNN_daily.csv","type":"text/csv","lastModified":1783243143355.151,"size":19992},{"id":"1b1ed7324f4f70c6dc809f221fc823edaedaba404339964929598ce5e0aecbb2a1b208236f52cd273a5344035a35e1151687d2bbd475254aa938894ab083e438","href":"https://static.observableusercontent.com/files/1b1ed7324f4f70c6dc809f221fc823edaedaba404339964929598ce5e0aecbb2a1b208236f52cd273a5344035a35e1151687d2bbd475254aa938894ab083e438","name":"VO2Max_daily.csv","type":"text/csv","lastModified":1783243143565.939,"size":2358},{"id":"288f77481abcaa9fd01b415d17a5a0fba954106ad2a69b09d8dcf2fe1cc57be8c053fcf7d80e5346000de8520769fd6a8904da8d102501ecd6d6cc131ccce019","href":"https://static.observableusercontent.com/files/288f77481abcaa9fd01b415d17a5a0fba954106ad2a69b09d8dcf2fe1cc57be8c053fcf7d80e5346000de8520769fd6a8904da8d102501ecd6d6cc131ccce019","name":"WalkingAsymmetryPercentage_daily.csv","type":"text/csv","lastModified":1783243143542.18,"size":30899},{"id":"03717589bf0139f5fe9de84970453a313cfc75c73247da309aa299c234d97fa791df0fe912eafe387ba484a3f94868a7b82d35e78c3ceab8586dfb4c7a4ef075","href":"https://static.observableusercontent.com/files/03717589bf0139f5fe9de84970453a313cfc75c73247da309aa299c234d97fa791df0fe912eafe387ba484a3f94868a7b82d35e78c3ceab8586dfb4c7a4ef075","name":"RestingHeartRate_daily.csv","type":"text/csv","lastModified":1783243143570.783,"size":12587}] 41:{"id":"76e9efc2ca04150f","slug":null,"title":"Pregnancy according to My Apple Watch","thumbnail":"f12c88a5c88c47788175793be723f0367ca92fca031533d45fc13bf26255067f","access_level":"public","create_time":"$D2026-07-05T09:17:26.025Z","update_time":"$D2026-07-07T14:18:09.808Z","view_time":null,"first_public_time":"$D2026-07-05T11:53:28.608Z","has_unpublished_edits":null,"fork":false,"stars":4,"starred":false,"owner":{"type":"team","id":"0f3922312d37021f","login":"sangyu-xu","name":"Sangyu Xu","avatar_url":"https://avatars-next.observableusercontent.com/147b7271ab77438aaa8c33ca080a47448ec8541791a3852921e7afd168ecba8a","tier":"starter_2024","bio":""},"version":240,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"bd8eaa52a9700eef","login":"sangyu","name":"Sangyu Xu","avatar_url":"https://avatars-next.observableusercontent.com/147b7271ab77438aaa8c33ca080a47448ec8541791a3852921e7afd168ecba8a","tier":null,"bio":""}],"members":[{"type":"individual","id":"bd8eaa52a9700eef","login":"sangyu","name":"Sangyu Xu","avatar_url":"https://avatars-next.observableusercontent.com/147b7271ab77438aaa8c33ca080a47448ec8541791a3852921e7afd168ecba8a","tier":null,"role":"editor"},{"type":"team","id":"0f3922312d37021f","login":"sangyu-xu","name":"Sangyu Xu","avatar_url":"https://avatars-next.observableusercontent.com/147b7271ab77438aaa8c33ca080a47448ec8541791a3852921e7afd168ecba8a","tier":"starter_2024","role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"titling":"auto","title":"Pregnancy according to My Apple Watch","cells":[{"id":9,"value":"$43","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":0,"value":"$44","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":13,"value":"d3 = require(\"d3@7\")\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":15,"value":"Plot = require(\"@observablehq/plot@0.6\")\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":17,"value":"Plotly = require(\"plotly.js-dist-min@2\")\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":19,"value":"Inputs = require(\"@observablehq/inputs@0.12\")\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":21,"value":"rawTables = ({\n VO2Max: await FileAttachment(\"VO2Max_daily.csv\").csv(),\n RestingHeartRate: await FileAttachment(\"RestingHeartRate_daily.csv\").csv(),\n HeartRateVariabilitySDNN: await FileAttachment(\"HeartRateVariabilitySDNN_daily.csv\").csv(),\n WalkingAsymmetryPercentage: await FileAttachment(\"WalkingAsymmetryPercentage_daily.csv\").csv(),\n WalkingDoubleSupportPercentage: await FileAttachment(\"WalkingDoubleSupportPercentage_daily.csv\").csv()\n})\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":23,"value":"parseDate = d3.utcParse(\"%Y-%m-%d\")\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":25,"value":"series = Object.fromEntries(\n Object.entries(rawTables).map(([metric, rows]) => [\n metric,\n rows\n .map((row) => ({date: parseDate(row.date), value: Number(row[metric])}))\n .filter((d) => d.date && Number.isFinite(d.value))\n .sort((a, b) => d3.ascending(a.date, b.date))\n ])\n)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":27,"value":"phaseOrder = [\n \"Pre-pregnancy\",\n \"Trimester 1\",\n \"Trimester 2\",\n \"Trimester 3\",\n \"Postpartum\",\n \"Training recovery\",\n \"Weaning\"\n]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":29,"value":"colors = new Map([\n [\"Pre-pregnancy\", \"#1a8a3c\"],\n [\"Trimester 1\", \"#D4B800\"],\n [\"Trimester 2\", \"#FFBA44\"],\n [\"Trimester 3\", \"#E26436\"],\n [\"Postpartum\", \"#8B2500\"],\n [\"Training recovery\", \"#7154D2\"],\n [\"Weaning\", \"#B430A9\"]\n])","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":31,"value":"shortPhase = new Map([\n [\"Pre-pregnancy\", \"Pre-preg\"],\n [\"Trimester 1\", \"T1\"],\n [\"Trimester 2\", \"T2\"],\n [\"Trimester 3\", \"T3\"],\n [\"Postpartum\", \"Postpartum\"],\n [\"Training recovery\", \"Training\"],\n [\"Weaning\", \"Weaning\"]\n])","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":33,"value":"phaseSpans = [\n [\"Pre-pregnancy\", \"2000-01-01\", \"2024-05-22\"],\n [\"Trimester 1\", \"2024-05-22\", \"2024-08-28\"],\n [\"Trimester 2\", \"2024-08-28\", \"2024-12-04\"],\n [\"Trimester 3\", \"2024-12-04\", \"2025-02-28\"],\n [\"Postpartum\", \"2025-02-28\", \"2026-02-22\"],\n [\"Training recovery\", \"2026-02-22\", \"2026-06-01\"],\n [\"Weaning\", \"2026-06-01\", \"2100-01-01\"]\n].map(([phase, start, end]) => ({\n phase,\n start: parseDate(start),\n end: parseDate(end),\n color: colors.get(phase),\n short: shortPhase.get(phase)\n}))","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":35,"value":"keyEvents = [\n [\"2023-09-15\", \"VO2 peak\"],\n [\"2023-10-24\", \"Transatlantic Trip\"],\n [\"2023-12-14\", \"COVID\"],\n [\"2025-02-18\", \"Birth\"],\n [\"2025-04-24\", \"VO2 floor\"],\n [\"2026-02-22\", \"Training starts\"],\n [\"2026-06-01\", \"Weaning\"]\n].map(([date, label]) => ({date: parseDate(date), label}))","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":37,"value":"function getPhase(date) {\n if (date < parseDate(\"2024-05-22\")) return \"Pre-pregnancy\";\n if (date < parseDate(\"2024-08-28\")) return \"Trimester 1\";\n if (date < parseDate(\"2024-12-04\")) return \"Trimester 2\";\n if (date < parseDate(\"2025-02-18\")) return \"Trimester 3\";\n if (date < parseDate(\"2026-02-22\")) return \"Postpartum\";\n if (date < parseDate(\"2026-06-01\")) return \"Training recovery\";\n return \"Weaning\";\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":39,"value":"dayKey = (date) => d3.utcFormat(\"%Y-%m-%d\")(date)\n","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":41,"value":"function hammingSmooth(values, window = 30) {\n if (window < 3 || values.length < window) return values.slice();\n\n const weights = Array.from({length: window}, (_, i) => (\n 0.54 - 0.46 * Math.cos((2 * Math.PI * i) / (window - 1))\n ));\n const total = d3.sum(weights);\n const normalized = weights.map((d) => d / total);\n const pad = Math.floor(window / 2);\n\n const reflect = (index) => {\n if (index < 0) return -index;\n if (index >= values.length) return 2 * values.length - index - 2;\n return index;\n };\n\n return values.map((_, i) => d3.sum(normalized, (weight, j) => (\n values[reflect(i - pad + j)] * weight\n )));\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":43,"value":"$45","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":45,"value":"coverage = await FileAttachment(\"health_metric_coverage.csv\").csv()\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":47,"value":"function coverageTable(rows) {\n const columns = [\"Metric\", \"Unit\", \"Obs (days)\", \"Coverage\", \"From\", \"To\"];\n const table = document.createElement(\"table\");\n table.className = \"coverage\";\n table.appendChild(Object.assign(document.createElement(\"thead\"), {\n innerHTML: `${columns.map((c) => `${c}`).join(\"\")}`\n }));\n const tbody = document.createElement(\"tbody\");\n for (const row of rows) {\n const tr = document.createElement(\"tr\");\n for (const column of columns) {\n const td = document.createElement(\"td\");\n td.textContent = row[column];\n tr.appendChild(td);\n }\n tbody.appendChild(tr);\n }\n table.appendChild(tbody);\n return table;\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":49,"value":"md`## Data Coverage`\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":55,"value":"coverageTable(coverage)\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":57,"value":"md`## Time Series\n30-day Hamming-smoothed trends for the three state-space axes, plus walking double support as the comparison signal.`\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":64,"value":"showMetrics = [\n [\"VO2Max\", \"mL/kg/min\"],\n [\"RestingHeartRate\", \"bpm\"],\n [\"WalkingAsymmetryPercentage\", \"proportion\"],\n [\"WalkingDoubleSupportPercentage\", \"proportion\"]\n].filter(([metric]) => series[metric]?.length)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":66,"value":"$46","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":68,"value":"timeSeriesGrid = {\n return function(width) {\n const grid = document.createElement(\"div\");\n grid.className = \"metric-grid\";\n\n for (const [metric, unit] of showMetrics) {\n const panel = document.createElement(\"section\");\n panel.className = \"metric-panel\";\n\n const title = document.createElement(\"h3\");\n title.className = \"metric-title\";\n title.textContent = metric;\n\n panel.append(title, plotMetric(metric, unit, Math.max(320, width)));\n grid.appendChild(panel);\n }\n\n return grid;\n };\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":141,"value":"key_metrics_timeseries = {\n return timeSeriesGrid(width);\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":70,"value":"timeSeriesGrid(width)\n","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":72,"value":"md`## 3D Physiological State Space\n\n**Main:** VO2Max x Resting HR x Walk Asymmetry. \n**Comparison:** same x/y axes with Walking Double Support Percentage as z.`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":74,"value":"function buildStateSpace(zMetric, zColumn) {\n const parts = new Map([\n [\"VO2Max\", smoothSeries(series.VO2Max, 30)],\n [\"RestingHR\", smoothSeries(series.RestingHeartRate, 30)],\n [zColumn, smoothSeries(series[zMetric], 30)]\n ]);\n\n const maps = new Map(Array.from(parts, ([name, rows]) => [\n name,\n new Map(rows.map((d) => [dayKey(d.date), d.value]))\n ]));\n\n const keys = Array.from(maps.get(\"VO2Max\").keys())\n .filter((key) => maps.get(\"RestingHR\").has(key) && maps.get(zColumn).has(key));\n\n return keys.map((key) => {\n const date = parseDate(key);\n return {\n date,\n dateLabel: key,\n VO2Max: maps.get(\"VO2Max\").get(key),\n RestingHR: maps.get(\"RestingHR\").get(key),\n [zColumn]: maps.get(zColumn).get(key),\n phase: getPhase(date)\n };\n });\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":76,"value":"ssAsym = buildStateSpace(\"WalkingAsymmetryPercentage\", \"WalkAsym\")\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":78,"value":"ssDbl = buildStateSpace(\"WalkingDoubleSupportPercentage\", \"DoubleSup\")","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":80,"value":"key3dAsym = [\n [\"2023-06-14\", \"VO2 peak
38.6\", \"#1a8a3c\"],\n [\"2025-02-18\", \"Birth\", \"#8B2500\"],\n [\"2025-04-24\", \"VO2 floor
21.2\", \"#8B2500\"],\n [\"2026-02-22\", \"Training
starts\", \"#7154D2\"],\n [\"2026-07-01\", \"Now\", \"#4954D6\"]\n].map(([date, label, color]) => ({date: parseDate(date), label, color}))\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":82,"value":"key3dDbl = key3dAsym","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":84,"value":"nearestRow = (rows, date) => d3.least(rows, (row) => Math.abs(row.date - date))\n","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":88,"value":"$47","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":104,"value":"\nmd`## Animated trajectory\n\nPress Play to watch the state evolve from 2023 to 2026. The dot colour\nfollows the current life phase. The full grey trajectory is shown as context.`\n","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":100,"value":"$48","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":92,"value":"md`## Interpretation\n\nThis is what a recovery in progress looks like in the data. The metabolic signals are trending back. The biomechanical ones aren't there yet. `\n","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W49","stdlib":"1"}} 42:"$41:body"