Understanding the bias in machine learning systems for cardiovascular disease risk assessment

Artificial Intelligence (AI), in particular, machine learning (ML) has shown promising results in coronary artery disease (CAD) or cardiovascular disease (CVD) risk prediction. Bias in ML systems is of great interest due to its over-performance and poor clinical delivery. The main objective is to un...

Teljes leírás

Elmentve itt :
Bibliográfiai részletek
Szerzők: Suri Jasjit S.
Bhagawati Mrinalini
Paul Sudip
Protogeron Athanasios
Sfikakis Petros P.
Kitas George D.
Khanna Narendra N.
Ruzsa Zoltán
Sharma Aditya M.
Saxena Sanjay
Faa Gavino
Paraskevas Kosmas I.
Laird John R.
Johri Amer M.
Saba Luca
Kalra Manudeep
Dokumentumtípus: Cikk
Megjelent: 2022
Sorozat:Computers in biology and medicine 142
Tárgyszavak:
doi:10.1016/j.compbiomed.2021.105204

mtmt:32611796
Online Access:http://publicatio.bibl.u-szeged.hu/23378
LEADER 02925nab a2200397 i 4500
001 publ23378
005 20220131145625.0
008 220131s2022 hu o 0|| Angol d
022 |a 1879-0534 
024 7 |a 10.1016/j.compbiomed.2021.105204  |2 doi 
024 7 |a 32611796  |2 mtmt 
040 |a SZTE Publicatio Repozitórium  |b hun 
041 |a Angol 
100 1 |a Suri Jasjit S. 
245 1 0 |a Understanding the bias in machine learning systems for cardiovascular disease risk assessment  |h [elektronikus dokumentum] /  |c  Suri Jasjit S. 
260 |c 2022 
300 |a Terjedelem: 23 p-Azonosító: 105204 
490 0 |a Computers in biology and medicine  |v 142 
520 3 |a Artificial Intelligence (AI), in particular, machine learning (ML) has shown promising results in coronary artery disease (CAD) or cardiovascular disease (CVD) risk prediction. Bias in ML systems is of great interest due to its over-performance and poor clinical delivery. The main objective is to understand the nature of risk-of-bias (RoB) in ML and non-ML studies for CVD risk prediction.PRISMA model was used to shortlisting 117 studies, which were analyzed to understand the RoB in ML and non-ML using 46 and 32 attributes, respectively. The mean score for each study was computed and then ranked into three ML and non-ML bias categories, namely low-bias (LB), moderate-bias (MB), and high-bias (HB), derived using two cutoffs. Further, bias computation was validated using the analytical slope method.Five types of the gold standard were identified in the ML design for CAD/CVD risk prediction. The low-moderate and moderate-high bias cutoffs for 24 ML studies (5, 10, and 9 studies for each LB, MB, and HB) and 14 non-ML (3, 4, and 7 studies for each LB, MB, and HB) were in the range of 1.5 to 1.95. BiasML< Biasnon-ML by ∼43%. A set of recommendations were proposed for lowering RoB.ML showed a lower bias compared to non-ML. For a robust ML-based CAD/CVD prediction design, it is vital to have (i) stronger outcomes like death or CAC score or coronary artery stenosis; (ii) ensuring scientific/clinical validation; (iii) adaptation of multiethnic groups while practicing unseen AI; (iv) amalgamation of conventional, laboratory, image-based and medication-based biomarkers. 
650 4 |a Klinikai orvostan 
700 0 1 |a Bhagawati Mrinalini  |e aut 
700 0 1 |a Paul Sudip  |e aut 
700 0 1 |a Protogeron Athanasios  |e aut 
700 0 1 |a Sfikakis Petros P.  |e aut 
700 0 1 |a Kitas George D.  |e aut 
700 0 1 |a Khanna Narendra N.  |e aut 
700 0 1 |a Ruzsa Zoltán  |e aut 
700 0 1 |a Sharma Aditya M.  |e aut 
700 0 1 |a Saxena Sanjay  |e aut 
700 0 1 |a Faa Gavino  |e aut 
700 0 1 |a Paraskevas Kosmas I.  |e aut 
700 0 1 |a Laird John R.  |e aut 
700 0 1 |a Johri Amer M.  |e aut 
700 0 1 |a Saba Luca  |e aut 
700 0 1 |a Kalra Manudeep  |e aut 
856 4 0 |u http://publicatio.bibl.u-szeged.hu/23378/1/SuriCompBiolMed2022.pdf  |z Dokumentum-elérés