Causality Refined Diagnostic Prediction

Applying machine learning in the health care domain has shown promising results in recent years. Interpretable outputs from learning algorithms are desirable for decision making by health care personnel. In this work, we explore the possibility of utilizing causal relationships to refine diagnostic prediction. We focus on the task of diagnostic prediction using discomfort drawings, and explore two ways to employ causal identification to improve the diagnostic results. Firstly, we use causal identification to infer the causal relationships among diagnostic labels which, by itself, provides interpretable results to aid the decision making and training of health care personnel. Secondly, we suggest a post-processing approach where the inferred causal relationships are used to refine the prediction accuracy of a multi-view probabilistic model. Experimental results show firstly that causal identification is capable of detecting the causal relationships among diagnostic labels correctly, and secondly that there is potential for improving pain diagnostics prediction accuracy using the causal relationships.

[1]  Peter Spirtes,et al.  Causal discovery and inference: concepts and recent methodological advances , 2016, Applied Informatics.

[2]  Hedvig Kjellström,et al.  Inter-battery Topic Representation Learning , 2016, ECCV.

[3]  L. Strender,et al.  Reliability of Clinical Tests in the Assessment of Patients With Neck/Shoulder Problems—Impact of History , 2003, Spine.

[4]  Peter Spirtes,et al.  Introduction to Causal Inference , 2010, J. Mach. Learn. Res..

[5]  Yasuhisa Tanaka,et al.  Cervical Roots as Origin of Pain in the Neck or Scapular Regions , 2006, Spine.

[6]  H. Palmer Pain charts; a description of a technique whereby functional pain may be diagnosed from organic pain. , 1949, The New Zealand medical journal.

[7]  Lars-Erik Strender,et al.  Pain drawing in the assessment of neurogenic pain and dysfunction in the neck/shoulder region: inter-examiner reliability and concordance with clinical examination. , 2007, Pain medicine.

[8]  J. Pearl Causality: Models, Reasoning and Inference , 2000 .

[9]  M. Albeck A critical assessment of clinical diagnosis of disc herniation in patients with monoradicular sciatica , 2005, Acta Neurochirurgica.

[10]  Nir Friedman,et al.  Probabilistic Graphical Models - Principles and Techniques , 2009 .

[11]  Byron C. Wallace,et al.  Diagnostic Prediction Using Discomfort Drawings with IBTM , 2016 .

[12]  L. Strender,et al.  Assessment of nerve involvement in the lumbar spine: agreement between magnetic resonance imaging, physical examination and pain drawing findings , 2010, BMC musculoskeletal disorders.

[13]  Richard Scheines,et al.  Causation, Prediction, and Search, Second Edition , 2000, Adaptive computation and machine learning.

[14]  Pain and pathology in lumbar disc hernia. , 1995, Clinical orthopaedics and related research.

[15]  J. Ekholm,et al.  Relation between pain location and disc pathology: a study of pain drawings and CT/discography. , 1999, The Clinical journal of pain.