Welcome to BIBM Workshop
The Third Workshop on Machine Learning and Big Data Analysis for Disease Classification
| Disease classification is a fundamental problem in diagnostics, genetic association, and treatment matching and personalization. Refinement of disease classification can lead to customized treatment for a complex disease. With the advances and fast development in machine learning and big data techniques, there is great progress in disease classification. Different from general classification, disease classification has a variety of problems to deal with, such as missing values, heterogeneity across different data sources, the need to factor in biological knowledge and medical knowledge of a disorder and so on. |
| Novel methods, statistical models and software systems are needed to address the challenges in disease classification and phenotyping. Classic methods may not achieve the analytic goal in this area. For instance, multiple imputation may be insufficient to deal with the missing values that mix obligated missing and random missing. Obligated-missing entries in a survey instrument actually encode important diagnostic information. The different data modalities used in disease classification impose additional challenges. Sophisticated transfer learning, domain adaptation, multi-task learning, multi-view data analytics might be feasible solutions. Additional caution may also be necessary in modeling temporal or spatial structures in the data, and in coping with the massive sample size and data dimensions. |
| There are two general lines of research for disease classification: classifying by clinical manifestations or by etiology. Clinical classifications are often useful for treatment and management. Etiologic classifications can be more useful for prevention. For both methods, phenotyping is very important to characterize and represent the disease. Up to date, there is great progress in disease data acquisition and collection as well as in the development of machine learning and bioinformatics methods, which create a dedicated subarea to health care. This workshop aims to provide a forum for academic and industrial researchers and physicians to exchange research ideas/designs and share research findings to promote the development of refined classification of complex disorders. |
|
Topics In this workshop, we solicit papers that cover but are not limited to the following topics.
|
| Submission
Your paper should be formatted to IEEE Computer Society Proceedings Manuscript Formatting Guidelines Although we accept submissions in the form of PDF, PS, and DOC/RTF files, you are strongly encouraged to generate a PDF version for your paper submission if your paper was prepared in Word. Please submit your paper at Machine Learning and Big Data Research for Disease Classification and Complex Phenotyping All papers accepted for workshops will be included in the Workshop Proceedings published by the IEEE Computer Society Press, which are indexed by EI. |
| Presentation schedule
9:10-9:15 Predicting Sentinel Node Status in Melanoma from a Real-World EHR Dataset.
|
| Organizing Chairs
Jinbo Bi, Associate Professor of Computer Science and Engineering Department, University of Connecticut |
| Program Committee
Le Lu, Staff Scientist, Department of Clinical Image Processing Service, National Institutes of Health |
| Important Dates
Oct 11, 2017: Due date for full workshop papers submission |
The first workshop we organized can be found in: Bioinformatics on Subtyping of Complex Phenotypes and Genetic Analysis of Phenotypic Subtypes.
The second workshop we organized can be found in :Machine Learning and Big Data Research for Disease Classification and Complex Phenotyping.