Decision Support System for Diagnosis and Treatment of Hearing Disorders

The book presents a knowledge discovery based approach to build a recommender system supporting a physician in treating tinnitus patients with the highly successful method called Tinnitus Retraining Therapy.
It describes experiments on extracting novel knowledge from the historical dataset of patients treated by Dr. P. Jastreboff so that to better understand factors behind therapy's effectiveness and better personalize treatments for different profiles of patients.
The book is a response for a growing demand of an advanced data analytics in the healthcare industry in order to provide better care with the data driven decision-making solutions.
The potential economic benefits of applying computerized clinical decision support systems include not only improved efficiency in health care delivery (by reducing costs, improving quality of care and patient safety), but also enhancement in treatment's standardization, objectivity and availability in places of scarce expert's knowledge on this difficult to treat hearing disorder.
Furthermore, described approach could be used in assessment of the clinical effectiveness of evidence-based intervention of various proposed treatments for tinnitus.

Describes a recommender system for tinnitus to help physicians diagnose and treat tinnitus patients Presents an introduction to the topic of tinnitus Highlights the basic concepts of recommender systems (RS), the current state-of-the-art and their real-world applications in different areas Focuses on health recommender systems Proposes a knowledge discovery approach to developing decision support systems Presents theoretical concepts and algorithms for rule-based systems, including: decision tables, classification rules, action rule extraction, and meta-actions Includes supplementary material: sn.pub/extras

Inhalt
Preface.- Introduction.- Tinnitus treatment as a problem area .- Recommender solutions overview.- Knowledge discovery approach for recommendation.- RECTIN system design.- Experiment 1: classiers.- Experiment 2: diagnostic rules.- Experiment 3: treatment rules.- Experiment 4: treatment rules enhancement.- RECTIN implementation.- Final conclusions and future work.- References.