{
  "docId": "019dd923-5e88-73ef-bd5d-06b04d219fea",
  "docSlug": "dd91c78f6570bf29",
  "documentTitle": "2023 Air Street Capital The State of AI Report 2023",
  "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": 66,
  "pageCount": 163,
  "prevPage": 65,
  "nextPage": 67,
  "slideType": "case_study",
  "function": "illustrate_case",
  "density": "dense",
  "nDataPoints": 2,
  "notes": "The slide presents a specific technical architecture (CoDoC) for AI-human decision deferral in clinical settings.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "process_diagram",
    "area_chart"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-06b04d219fea/66",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea#slide-66",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "For breast cancer screening, CoDoC reduces false positives by 25% at the same false-negative rate compared to double reading with arbitration in the UK. Importantly, clinical workload is reduced by 66% as a result.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a226-7e7c68378e96",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.55,
        "w": 0.4,
        "x": 0.55,
        "y": 0.35
      },
      "kind": "image",
      "text": "Diagrams showing CoDoC deployment architecture, training method, and deferral regions.",
      "attrs": null,
      "subkind": "infographic",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "d2a33ff2-f7bb-468d-8bae-6c8772fe6857",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.3,
        "w": 0.45,
        "x": 0.02,
        "y": 0.38
      },
      "kind": "list",
      "text": "Complementarity-Driven Deferral to Clinical Workflow (CoDoC) learns to decide whether to rely on a predictive AI model's output or defer to a clinical workflow instead.\nFor breast cancer screening, CoDoC reduces false positives by 25% at the same false-negative rate compared to double reading with arbitration in the UK. Importantly, clinical workload is reduced by 66% as a result.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "e929fe19-ee6b-40ab-9482-120169caccc1",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "clinical workload reduction: 66%",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a226-816349a18e62",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.15,
        "w": 0.45,
        "x": 0.02,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "Computer vision has been shown to be useful for breast cancer screening on mammograms and tuberculosis triaging. However, to enable practical and reliable use in the clinic it is important to know when to rely on a predictive AI model or revert to a clinical workflow.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "f88acd5d-c775-4520-bb26-f11d6783567c",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.8,
        "x": 0.02,
        "y": 0.14
      },
      "kind": "title",
      "text": "Real world-inspired clinical system design for automated medical image analysis",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "84223706-89dd-4528-adf4-994d9c1de1bf",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Problem-Agitate-Solve (PAS)",
      "slug": "problem-agitate-solve-pas",
      "agent": "Storyteller",
      "layer": "slide",
      "matchId": "d08b841c-3970-41af-b5e8-dfd9fa40001d",
      "evidence": "Complementarity-Driven Deferral to Clinical Workflow (CoDoC) learns to decide whether to rely on a predictive AI model's output or defer to a clinical workflow instead.",
      "confidence": 0.6
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 120,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-41afd44ef59f",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Research + Industry sections inventory model, compute, funding facts.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 120,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-523bfb7f96e6",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Escalating capabilities, compute concentration and capital flows.",
      "position": 2,
      "confidence": 45,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 68,
      "from": 59,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-f9a066ea1e78",
      "evidence": "RFdiffusion, ESMFold, PerturbNet, AlphaMissense, Med-PaLM, PLIP, CoDoC, then meta map of AI-for-science fields.",
      "position": 7,
      "objective": "AI for science accumulates breakthroughs across bio and medicine",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 78,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}