AI meets MD: Combining Machine learning algorithms and medical experts’ diagnosis (MD) for new beginnings in the prediction of Alzheimer’s disease progression.

Abstract

Alzheimer's Disease (AD) is becoming a significant economic and social burden, yet the conventional process of medical diagnosis (MD) evinces significant proneness to error. One way of improving diagnostic accuracy may lie in data analysis performed by machine learning algorithms (MLA). However, limited availability of longitudinal biomarker information of AD progression leads to a data imbalance problem. This issue may be alleviated by designing Clinical Decision Support Systems (CDSS) incorporating both algorithmic and clinical predictions. Thus, the aim of the current study was to investigate the concordance between the assessment of progression predictors by clinicians and MLA. A survey including plots depicting sampling distributions of the most relevant predictors for three AD progression groups was created. Nine clinicians were contacted and asked to assign each patient to one of the groups. None of the clinicians fully completed the survey and only one provided detailed feedback. The results are interpreted as indicative of health professionals’ scepticism towards MLA.

Please cite this contribution as follows:

Bruton, O., Jari, S., & Andrade Leon, R. (2021). AI meets MD: Combining Machine learning algorithms and medical experts’ diagnosis (MD) for new beginnings in the prediction of Alzheimer’s disease progression. "forsch!" - Studentisches Online-Journal der Universität Oldenburg, 1.

https://nbn-resolving.org/urn:nbn:de:101:1-2023101019425861200859