{
  "docId": "019dd923-5e88-73ef-bd5c-fd5012384ae3",
  "docSlug": "bf350dd574c19997",
  "documentTitle": "2021 Air Street Capital The State of AI Report 2021",
  "authorId": "AirStreetCapital",
  "authorName": "Air Street Capital",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "vc_research",
  "sourceTypeLabel": "VC research",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.777,
  "pageNumber": 52,
  "pageCount": 188,
  "prevPage": 51,
  "nextPage": 53,
  "slideType": "industry_trends",
  "function": "analyze_data",
  "density": "dense",
  "nDataPoints": 3,
  "notes": "The slide highlights the lack of geographic diversity in training data for medical AI, leading to potential clinical bias.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "map"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5c-fd5012384ae3/52",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5c-fd5012384ae3#slide-52",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Thirty four states did not contribute data, point to huge patient underrepresentation.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c3-7407-8d28-dc5ebdc8b53d",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.43,
        "x": 0.52,
        "y": 0.42
      },
      "kind": "chart",
      "text": "Geographic distribution of US cohorts used to train deep learning algorithms",
      "attrs": null,
      "subkind": "map",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "748474ad-e07e-4285-90d1-271e43a04713",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.35,
        "w": 0.45,
        "x": 0.05,
        "y": 0.48
      },
      "kind": "list",
      "text": "Large proportions of the US population have been excluded from medical AI training data sets in radiology, ophthalmology, dermatology, pathology, gastroenterology, and cardiology.\nAI models often perform poorly on populations that are not represented in the training data. It is critical for AI training data to mirror the populations for which model are ultimately serving.\nUnderrepresented populations might also have specific problems that remain unaddressed.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "4b81dcb8-91aa-4be4-8b51-0ab567df6a20",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Patient cohort distribution: 71%",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c3-7407-8d28-e15a19307ccb",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.45,
        "x": 0.05,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "56 studies were published between 2015-19 that reported the training of a deep learning algorithm on at least one geographically identifiable patient cohort to perform an image-based diagnostic task vs. a human physician across 6 clinical disciplines. Of these studies, 71% used a patient cohort from one of three states: California, Massachusetts or New York. Thirty four states did not contribute data, point to huge patient underrepresentation.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "8d9d292c-16f0-4504-adb0-b383a5938dcd",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.43,
        "x": 0.52,
        "y": 0.86
      },
      "kind": "source-note",
      "text": "REBECCA ROBBINS/STAT SOURCE: \"GEOGRAPHIC DISTRIBUTION OF US COHORTS USED TO TRAIN DEEP LEARNING ALGORITHMS,\" JAMA 2020.",
      "attrs": null,
      "subkind": null,
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "4776bf5a-b80d-4f74-a939-8fa988cd543f",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.9,
        "x": 0.05,
        "y": 0.14
      },
      "kind": "title",
      "text": "Data deserts in biomedical AI research are likely to result in model bias in the clinic",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "0d7b2dca-3a48-4320-aed2-9b44f26af9e7",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 152,
      "from": 5,
      "beatId": "019dd95a-0682-776c-8e35-0affbadec38e",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Definitions p.5-6, Exec Summary p.7, then Sections 1-3 catalog evidence.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 152,
      "from": 10,
      "beatId": "019dd95a-0682-776c-8e35-1b3ee3a1c012",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Three sections accumulating evidence of accelerating progress.",
      "position": 2,
      "confidence": 55,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 55,
      "from": 52,
      "name": "Iceberg",
      "slug": "10-iceberg",
      "bestFor": "Consulting, complex problem solving, organizational change",
      "matchId": "019dd95a-07fe-70ce-8d3d-5a2841b69367",
      "evidence": "Data deserts -> measuring bias -> 94% systems worse than radiologist -> racial identification — symptom to system to root.",
      "position": 7,
      "objective": "Surface bias problems in medical AI",
      "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause",
      "confidence": 78,
      "description": "Reveal that the visible problem is merely a symptom of a deeper root cause"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}