2019/12/25 by Arun Verma, Manjesh K. Hanawal, Verma, Arun +3
Computer Science · #Applications (stat.AP) #Artificial Intelligence (cs.AI) #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Imbalanced Data Classification Techniques #Machine Learning (cs.LG) #Machine Learning in Healthcare
paper · pdf · doi:10.48550/arxiv.2001.00626
openalex publication_date 2019/12/25 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28
In medical diagnosis, physicians predict the state of a patient by checking\nmeasurements (features) obtained from a sequence of tests, e.g., blood test,\nurine test, followed by invasive tests. As tests are often costly, one would\nlike to obtain only those features (tests) that can establish the presence or\nabsence of the state conclusively. Another aspect of medical diagnosis is that\nwe are often faced with unsupervised prediction tasks as the true state of the\npatients may not be known. Motivated by such medical diagnosis problems, we\nconsider a it Cost-Sensitive Medical Diagnosis (CSMD) problem, where the\ntrue state of patients is unknown. We formulate the CSMD problem as a feature\nselection problem where each test gives a feature that can be used in a\nprediction model. Our objective is to learn strategies for selecting the\nfeatures that give the best trade-off between accuracy and costs. We exploit\nthe `Weak Dominance' property of problem to develop online algorithms that\nidentify a set of features which provides an `optimal' trade-off between cost\nand accuracy of prediction without requiring to know the true state of the\nmedical condition. Our empirical results validate the performance of our\nalgorithms on problem instances generated from real-world datasets.\n