1
play

1 Differential Diagnosis March 14, 2019 Diagnosis is the - PowerPoint PPT Presentation

Oct 01, 2023 •598 likes •1.12k views

1 Differential Diagnosis March 14, 2019 Diagnosis is the identification of the nature and cause of a certain phenomenon di ff erential diagnosis is the distinguishing of a particular disease or condition from others that

  1. � 1

  2. Differential Diagnosis 𝜌 March 14, 2019 “ Diagnosis is the identification of the nature and cause of a certain phenomenon” “ di ff erential diagnosis is the distinguishing of a particular disease or condition from others that present similar clinical features” —Wikipedia � 2

  3. Guyton's Model of Cardio- vascular Dynamics � 3

  4. Models for Diagnostic Reasoning • Flowcharts • Based on associations between diseases and {signs, symptoms} • “manifestations” covers all observables, including lab tests, bedside measurements, … • Single disease vs. multiple diseases • Probabilistic vs. categorical • Utility theoretic • Rule-based Sign : Any objective evidence of disease, as • Pattern matching opposed to a symptom , which is, by nature, subjective. For example, gross blood in the stool is a sign of disease; it is evidence that can be recognized by the patient, physician, nurse, or someone else. Abdominal pain is a symptom; it is something only the patient can perceive. https://www.medicinenet.com/script/main/art.asp? articlekey=5493 � 4

  5. Flowchart • BI/Lincoln Labs Clinical Protocols � 5

  6. Disease = {signs & symptoms} s1 s1 s2 s2 s3 s3 s4 s4 s5 s5 s6 s6 s7 s7 s8 s8 Disease Disease s9 s9 s10 s10 s... s... � 6

  7. Diagnosis by Card Selection s1 s1 s2 s1 s2 s1 s3 s2 s3 s2 s4 s3 s4 s3 s5 s4 s5 s4 s6 s5 s6 s5 s7 s6 s7 s6 s8 Disease s7 s8 Disease s7 s9 s8 Disease s9 s8 s10 Disease s9 s10 s9 s... s10 s... s10 s... s... � 7

  8. Naïve Bayes • Exhaustive and Mutually Exclusive disease M1 hypotheses (1 and only 1) • Conditionally independent observables (manifestations) M2 • P(D i ), P(M ij |D i ) D M3 M4 M5 M6 � 8

  9. How certain are we after a test? Imagine P(D+) = .001 (it’s a rare disease) T+ Accuracy of test = P(T+|D+) = P(T-|D-) = . 95 TP=p(T+|D+) D+ FN=p(T-|D+) p(D+) T- D? T+ p(D-)=1-p(D+) FP=p(T+|D-) D- Bayes’ Rule: TN=p(T-|D-) T- � 9

  10. Diagnostic Reasoning with Naive Bayes • Exploit assumption of conditional independence among symptoms • Sequence of observations of symptoms, S i , each revise the distribution via Bayes’ Rule D 1 : 0.12 D 1 : 0.19 D 1 : 0.08 D 1 : 0.01 obs S i obs S j obs S k D 2 : 0.37 D 2 : 0.30 D 2 : 0.59 D 2 : 0.96 ... ... ... ... D n : 0.03 D n : 0.01 D n : 0.05 D n : 0.00 • After the j-th observation, � 10

  11. Odds-Likelihood • In gambling, “3-to-1” odds means 75% chance of success • P = 0.5 means O=1 • Likelihood ratio • Odds-likelihood form of Bayes rule • Log transform � 11

  12. Test Thresholds - + FN FP T � 12

  13. Wonderful Test - + FN FP T � 13

  14. Test Thresholds Change Trade-off between Sensitivity and Specificity - + FN FP T � 14

  15. Receiver Operator Characteristic (ROC) Curve TPR (sensitivity) 1 T 0 1 0 FPR (1-specificity) � 15

  16. What makes a better test? TPR (sensitivity) superb 1 OK worthless 0 1 0 FPR (1-specificity) � 16

  17. Rationality • Every action has a cost • Principle of rationality • Act to maximize expected utility — homo economicus • Or minimize loss • Utility measures the value (“goodness”) of an outcome, e.g., • Life vs. death • Quality-adjusted life years (QALYs) � 17

  18. Case of a Man with Gangrene • From Pauker’s “Decision Analysis Service” at New England Medical Center Hospital, late 1970’s. • Man with gangrene of foot • Choose to amputate foot or treat medically • If medical treatment fails, patient may die or may have to amputate whole leg. • What to do? How to reason about it? � 18

  19. Decision Tree for Gangrene Case (Different sense of “Decision Tree” from ML/Classification!) Choice Chance 900 live (.99) 850 amputate foot 841.5 881 die (.01) 0 live (.98) 700 full recovery (.7) amputate leg 1000 medicine 686 die(.02) 0 871.5 worse (.25) live (.6) 686 995 medicine die (.05) 0 597 die (.4) 0 � 19

  20. “Folding back” a Decision Tree • The value of an outcome node is its utility • The value of a chance node is the expected value of its alternative branches; i.e., their values weighted by their probabilities • The value of a choice node is the maximum value of any of its branches � 20

  21. Where Do Utilities Come From? • Standard gamble • Would you prefer (choose one of the following two): 1. I chop o ff your foot 2. We play a game in which a fair process produces a random number r between 0 and 1 • If r > 0.8, I kill you; otherwise, you live on, healthy • If you’re indi ff erent, that’s the value of living without your foot! • I vary the 0.8 threshold until you are indi ff erent. • Alas, di ffi cult ascertainment problems! • Clearly depends on the individual • Not stable � 21

  22. Acute Renal Failure Program • Di ff erential Diagnosis of Acute Oliguric Renal Failure • “stop peeing” • 14 potential causes, exhaustive and mutually exclusive • 27 tests/questions/observations relevant to di ff erential • “cheap”; therefore, ordering based on expected information gain • 3 invasive tests (biopsy, retrograde pyelography, renal arteriography) • “expensive”; ordering based on (very naive) utility model • 8 treatments (conservative, IV fluids, surgery for obstruction, steroids, antibiotics, surgery for clots, antihypertensive drugs, heparin) • expected outcomes are “better”, “unchanged”, “worse” • Gorry, G. A., Kassirer, J. P ., Essig, A., & Schwartz, W. B. (1973). Decision analysis as the basis for computer-aided management of acute renal failure. The American Journal of Medicine , 55(3), 473–484.

Recommend

More recommend