Raymond R Bond1, Tomas Novotny2, Irena Andrsova2, Lumir Koc2, Martina Sisakova2, Dewar Finlay3, Daniel Guldenring3, James McLaughlin3, Aaron Peace4, Victoria McGilligan5, Stephen J Leslie6, Hui Wang3, Marek Malik7. 1. Faculty of Computing and Engineering, Ulster University, Newtownabbey, United Kingdom. Electronic address: rb.bond@ulster.ac.uk. 2. Department of Internal Medicine and Cardiology, University Hospital Brno and Faculty of Medicine of Masaryk University, Brno, Czechia. 3. Faculty of Computing and Engineering, Ulster University, Newtownabbey, United Kingdom. 4. Department of Cardiology, Altnagelvin Hospital, Western Health and Social Care Trust, United Kingdom. 5. Northern Ireland Centre for Stratified Medicine, C-TRIC, Ulster University, Derry~Londonderry, United Kingdom. 6. Cardiac Unit, Raigmore Hospital, Inverness, UK, IV2 3UJ. 7. St. Paul's Cardiac Electrophysiology and Imperial College London, London, United Kingdom.
Abstract
INTRODUCTION: Interpretation of the 12‑lead Electrocardiogram (ECG) is normally assisted with an automated diagnosis (AD), which can facilitate an 'automation bias' where interpreters can be anchored. In this paper, we studied, 1) the effect of an incorrect AD on interpretation accuracy and interpreter confidence (a proxy for uncertainty), and 2) whether confidence and other interpreter features can predict interpretation accuracy using machine learning. METHODS: This study analysed 9000 ECG interpretations from cardiology and non-cardiology fellows (CFs and non-CFs). One third of the ECGs involved no ADs, one third with ADs (half as incorrect) and one third had multiple ADs. Interpretations were scored and interpreter confidence was recorded for each interpretation and subsequently standardised using sigma scaling. Spearman coefficients were used for correlation analysis and C5.0 decision trees were used for predicting interpretation accuracy using basic interpreter features such as confidence, age, experience and designation. RESULTS: Interpretation accuracies achieved by CFs and non-CFs dropped by 43.20% and 58.95% respectively when an incorrect AD was presented (p < 0.001). Overall correlation between scaled confidence and interpretation accuracy was higher amongst CFs. However, correlation between confidence and interpretation accuracy decreased for both groups when an incorrect AD was presented. We found that an incorrect AD disturbs the reliability of interpreter confidence in predicting accuracy. An incorrect AD has a greater effect on the confidence of non-CFs (although this is not statistically significant it is close to the threshold, p = 0.065). The best C5.0 decision tree achieved an accuracy rate of 64.67% (p < 0.001), however this is only 6.56% greater than the no-information-rate. CONCLUSION: Incorrect ADs reduce the interpreter's diagnostic accuracy indicating an automation bias. Non-CFs tend to agree more with the ADs in comparison to CFs, hence less expert physicians are more effected by automation bias. Incorrect ADs reduce the interpreter's confidence and also reduces the predictive power of confidence for predicting accuracy (even more so for non-CFs). Whilst a statistically significant model was developed, it is difficult to predict interpretation accuracy using machine learning on basic features such as interpreter confidence, age, reader experience and designation.
INTRODUCTION: Interpretation of the 12‑lead Electrocardiogram (ECG) is normally assisted with an automated diagnosis (AD), which can facilitate an 'automation bias' where interpreters can be anchored. In this paper, we studied, 1) the effect of an incorrect AD on interpretation accuracy and interpreter confidence (a proxy for uncertainty), and 2) whether confidence and other interpreter features can predict interpretation accuracy using machine learning. METHODS: This study analysed 9000 ECG interpretations from cardiology and non-cardiology fellows (CFs and non-CFs). One third of the ECGs involved no ADs, one third with ADs (half as incorrect) and one third had multiple ADs. Interpretations were scored and interpreter confidence was recorded for each interpretation and subsequently standardised using sigma scaling. Spearman coefficients were used for correlation analysis and C5.0 decision trees were used for predicting interpretation accuracy using basic interpreter features such as confidence, age, experience and designation. RESULTS: Interpretation accuracies achieved by CFs and non-CFs dropped by 43.20% and 58.95% respectively when an incorrect AD was presented (p < 0.001). Overall correlation between scaled confidence and interpretation accuracy was higher amongst CFs. However, correlation between confidence and interpretation accuracy decreased for both groups when an incorrect AD was presented. We found that an incorrect AD disturbs the reliability of interpreter confidence in predicting accuracy. An incorrect AD has a greater effect on the confidence of non-CFs (although this is not statistically significant it is close to the threshold, p = 0.065). The best C5.0 decision tree achieved an accuracy rate of 64.67% (p < 0.001), however this is only 6.56% greater than the no-information-rate. CONCLUSION: Incorrect ADs reduce the interpreter's diagnostic accuracy indicating an automation bias. Non-CFs tend to agree more with the ADs in comparison to CFs, hence less expert physicians are more effected by automation bias. Incorrect ADs reduce the interpreter's confidence and also reduces the predictive power of confidence for predicting accuracy (even more so for non-CFs). Whilst a statistically significant model was developed, it is difficult to predict interpretation accuracy using machine learning on basic features such as interpreter confidence, age, reader experience and designation.
Authors: Tara A Retson; Kyle A Hasenstab; Seth J Kligerman; Kathleen E Jacobs; Andrew C Yen; Sharon S Brouha; Lewis D Hahn; Albert Hsiao Journal: Radiol Artif Intell Date: 2021-11-10
Authors: Navid Hasani; Faraz Farhadi; Michael A Morris; Moozhan Nikpanah; Arman Rhamim; Yanji Xu; Anne Pariser; Michael T Collins; Ronald M Summers; Elizabeth Jones; Eliot Siegel; Babak Saboury Journal: PET Clin Date: 2022-01
Authors: Claire M Felmingham; Nikki R Adler; Zongyuan Ge; Rachael L Morton; Monika Janda; Victoria J Mar Journal: Am J Clin Dermatol Date: 2021-03 Impact factor: 7.403
Authors: Rob Brisk; Raymond R Bond; Dewar Finlay; James A D McLaughlin; Alicja J Piadlo; David J McEneaney Journal: Front Physiol Date: 2022-03-17 Impact factor: 4.566
Authors: Ankita Bhat; Daria Podstawczyk; Brandon K Walther; John R Aggas; David Machado-Aranda; Kevin R Ward; Anthony Guiseppi-Elie Journal: J Transl Med Date: 2020-09-14 Impact factor: 5.531
Authors: Charles Richard Knoery; Janet Heaton; Rob Polson; Raymond Bond; Aleeha Iftikhar; Khaled Rjoob; Victoria McGilligan; Aaron Peace; Stephen James Leslie Journal: Crit Pathw Cardiol Date: 2020-09