One of the reasons I have always had a soft spot for IBM (not counting my favorable experiences with their SP systems), is the company’s ability to do some really cool science and technology development. It harks back to a day when some of the giants in the field (IBM, Intel, AT&T, etc) used to be at the forefront of innovation.
Recently, I found HealthMiner on the IBM web pages. What is HealthMiner? First of all, it is a middleware package, which makes it possible for IBM to make it available to their ISV partners. It integrates with other applications developed at IBM, namely Thoth (a pattern recognition tool), CliniMiner (a data mining method for finding hidden relationships in massive data sets), and Predictive Analysis (a decision support tool to develop classification tools). As the diagram below shows HealthMiner enables the processing and analysis of patient records to discover new rules and relationships that could lead to the development of novel classification schemes and identification of key factors responsible for a particular patient condition or response to a particular treatment, etc
One of the biggest challenges in coming years will be correlating patient data with molecular data. I theory, I suspect that HealthMiner should be applicable to mining correlations between clinical information and molecular information to point researchers towards new diagnostics and other interesting ways to classify diseases, etc.
When a behemoth does cool science
One of the reasons I have always had a soft spot for IBM (not counting my favorable experiences with their SP systems), is the company’s ability to do some really cool science and technology development. It harks back to a day when some of the giants in the field (IBM, Intel, AT&T, etc) used to be at the forefront of innovation.
Recently, I found HealthMiner on the IBM web pages. What is HealthMiner? First of all, it is a middleware package, which makes it possible for IBM to make it available to their ISV partners. It integrates with other applications developed at IBM, namely Thoth (a pattern recognition tool), CliniMiner (a data mining method for finding hidden relationships in massive data sets), and Predictive Analysis (a decision support tool to develop classification tools). As the diagram below shows HealthMiner enables the processing and analysis of patient records to discover new rules and relationships that could lead to the development of novel classification schemes and identification of key factors responsible for a particular patient condition or response to a particular treatment, etc
One of the biggest challenges in coming years will be correlating patient data with molecular data. I theory, I suspect that HealthMiner should be applicable to mining correlations between clinical information and molecular information to point researchers towards new diagnostics and other interesting ways to classify diseases, etc.
Technorati Tags: IBM, HealthMiner, Bioinformatics, Biomedical Informatics, Diagnostics