{
  "docId": "019dd923-5ca1-7489-b638-e91d3b8478cc",
  "docSlug": "9fbedbfea0231cbe",
  "documentTitle": "AI Healthcare Errors",
  "authorId": "McKinsey",
  "authorName": "McKinsey",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "strategy_consulting",
  "sourceTypeLabel": "Strategy consulting",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.333,
  "pageNumber": 11,
  "pageCount": 26,
  "prevPage": 10,
  "nextPage": 12,
  "slideType": "case_study",
  "function": "illustrate_case",
  "density": "dense",
  "nDataPoints": 7,
  "notes": "The slide uses a bar chart to compare the AUC of a random forest machine learning model against five established clinical heuristic scoring systems (CART, Logistic regression, CURB-65, MEDS, mREMS).",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "photo",
    "bar_chart_vertical",
    "footnote"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5ca1-7489-b638-e91d3b8478cc/11",
  "deckHref": "/decks/019dd923-5ca1-7489-b638-e91d3b8478cc",
  "deckJsonHref": "/decks/019dd923-5ca1-7489-b638-e91d3b8478cc.json",
  "deckAnchorHref": "/decks/019dd923-5ca1-7489-b638-e91d3b8478cc#slide-11",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "The machine learning approach has significantly greater predictive performance than other clinical decision rules based models.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd951-d3bc-73c4-a2b9-2f389ba03a8d",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.96,
        "x": 0.02,
        "y": 0.0065
      },
      "kind": "chart",
      "text": "Accuracy – AUC (area under the receiver operating characteristic curve)",
      "attrs": null,
      "subkind": "bar-vertical",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "08048f37-3b25-4242-8921-75f0ad15f500",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.46,
        "x": 0.52,
        "y": 0.00115
      },
      "kind": "image",
      "text": null,
      "attrs": null,
      "subkind": "photo",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "e7386375-4fa5-40fd-a83a-d514d9adf4d8",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.48,
        "x": 0.02,
        "y": 0.00115
      },
      "kind": "list",
      "text": "ER patients with sepsis have a 5% mortality risk; Machine learning (random forest model) can be used to accurately identify patients with sepsis in the ER using data from the Electronic Health Record; The machine learning approach has significantly greater predictive performance than other clinical decision rules based models; The method developed can be automated and applied to EHRs to predict other clinical outcomes of interest; Machine learning methods have potential advantages over traditional heuristic methods as they have greater generalizability, can be developed quickly, and may automatically update as new information becomes available",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "8b0e0395-b8cc-4bcb-845c-e98aa9144b6d",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "AUC (Area Under the Curve): 0.86",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd951-d3bc-73c4-a2b9-30dc327bb91a",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.06,
        "w": 0.96,
        "x": 0.02,
        "y": 0.92
      },
      "kind": "source-note",
      "text": "1 Standard statistical measure of diagnostic accuracy 2 CART = Classification and regression tree 3 Difference between the random forest model and all other models SOURCE: Taylor et al, Prediction of In-hospital Mortality in Emergency Department Patients With Sepsis: A Local Big Data-Driven, Machine Learning Approach, 2016, Academic Emergency Medicine, Volume 23, Issue 3, 269-278",
      "attrs": null,
      "subkind": null,
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "76ca65ee-04db-48c8-97c4-313281783cb0",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.96,
        "x": 0.02,
        "y": 0.03
      },
      "kind": "title",
      "text": "Proof of concept established for machine learning based predictive modelling to identify patients in ER with sepsis",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "5a430a05-5f6a-4fc7-87c8-e9a7698358b0",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [
    {
      "metricRaw": "AUC (Area Under the Curve)",
      "numberRaw": "0.86",
      "numberKind": "plain",
      "actionTitle": "Proof of concept established for machine learning based predictive modelling to identify patients in ER with sepsis",
      "calloutText": "The machine learning approach has significantly greater predictive performance than other clinical decision rules based models.",
      "numberScale": null,
      "numberValue": 0.86,
      "metricFamily": "other",
      "numberCurrency": null
    }
  ],
  "tools": [],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 26,
      "from": 10,
      "beatId": "9b6e4ffe-95ca-40d4-92d5-70f2cd8f60f8",
      "arcName": "Problem-Agitate-Solution",
      "arcSlug": "problem-agitate-solution",
      "beatName": "Solution (Provide relief)",
      "beatSlug": "problem-agitate-solution-solution-provide-relief",
      "evidence": "The deck offers AI as a solution to prevent healthcare errors and improve patient outcomes.",
      "position": 2,
      "confidence": 0.8,
      "parentBeatName": "Resolution",
      "parentBeatSlug": "resolution"
    }
  ],
  "loops": [
    {
      "to": 14,
      "from": 4,
      "name": "Golden Circle",
      "slug": "11-golden-circle",
      "bestFor": "Visionary leadership, brand positioning, mission statements",
      "matchId": "1b36e03c-871f-4ba9-977e-2461066df8bf",
      "evidence": "The deck presents various case studies and examples of AI applications in healthcare, using the golden circle framework.",
      "position": 1,
      "objective": "Explain why AI is different and how it can be applied in healthcare",
      "structure": "The Why (Belief) -> The How (Process) -> The What (Result)",
      "confidence": 0.6,
      "description": "Invert the typical pitch by starting with why you exist, rather than what you do"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}