 Business Science
Reduction of Post-operative
Length of Stay at Hospital
Assessing Carcinogenic
Potential of Chemicals
Indications for Treatment
of Duodenal Ulcer by HSV
Knowledge Discovery
from Clinical Databases
Predicting Reaction Rates
of Organic Compounds
from Spectral Data
Industry
|
|
|
These examples discuss application
of our tools or the underlying mathematical methods in scientific areas such as medical
diagnosis and cost containment. Because of confidentiality considerations, we
discuss here only applications that were made available to the public. Reduction of Post-operative
Length of Stay at Hospital
Assessing
Carcinogenic Potential of Chemicals
Indications for
Treatment of Duodenal Ulcer by HSV
Knowledge Discovery
from Clinical Databases
Predicting Reaction
Rates of Organic Compounds from Spectral Data
| Title: |
Reduction of post-operative length of stay at hospital. |
| Developer: |
SolveTech, CA |
| Description: |
The developer identified five independent variables that
determine the length of stay in the hospital for coronary bypass surgery. The variables
allowed adjustments/shortened length of hospital stay from 13 days to 5.5 days. |
| Requirements: |
- Detailed database of patients’ conditions and treatments.
|
| Benefits |
- Identification of control variables leading to reduction of patient post-operative stay
in hospital and associated cost by half.
- Identification of significant factors prolonging the length of stay.
|
 Top
| Title: |
Assessing Carcinogenic Potential of Chemicals |
| Developer: |
Health Designs Inc.
National Institute of Environmental Health Sciences
Research Triangle Park, NC, USA |
| Description: |
The developer proposed a model for predicting National
Toxicology Program calls for subchronic and chronic chemicals. |
| Requirements: |
- Large databases of toxicity data from bioassays.
|
| Benefits |
- Accurate rules (85 to 95 percent) for identification of positive and negative potential
for carcinogenic activity of untested chemicals.
- Discovery that evaluation of organ-specific subchronic data can be useful in predicting
carcinogenicity calls.
|
Top
| Title: |
Indications for Treatment of Duodenal Ulcer by HSV |
| Developer: |
Department of Surgery
F. Raszeja Hospital
Pozana, Poland |
| Description: |
The institute developed a method for classification of
patients with duodenal ulcer treated by highly selective vagotomy (HSV). Each patient was
described by eleven variables concerning anamnesis data and preoperative gastric
secretion. The developed rules formally confirm indications used by clinicians for HSV. |
| Requirements: |
- Clinical database containing anamnesis data and preoperative gastric secretion data.
|
| Benefits |
- A method for classification of patients for treatment of duodenal ulcer using HSV.
- Safer treatment of duodenal ulcer with lower frequency of complications and lower
mortality.
|
Top
| Title: |
Knowledge Discovery from Clinical Databases |
| Developer: |
Medical Research Institute
Tokyo Medical and Dental University
Tokyo, Japan |
| Description: |
The developer compares application of medical diagnosis rule
induction methods for discerning expert knowledge from medical databases. The comparison
shows advantages of these methods in situations where medical databases contain
information of both deterministic and probabilistic nature. |
| Requirements: |
- Medical databases containing patients' diagnoses and symptoms.
|
| Benefits |
- High accuracy diagnostic rules for headache, meningitis and CVD medical cases
classification.
|
Top
| Title: |
Predicting Reaction Rates of Organic Compounds from
Spectral Data |
| Developer: |
US Environmental Protection Agency
Athens, Georgia, USA |
| Description: |
The developer applied analysis of spectroscopic-based data as
a means for predicting pollutant transport and transformation parameters for use in
environmental risk assessment. One of the issues addressed was development of a method for
ranking the predictive power of the individual interferometric points. |
| Requirements: |
- Databases of hydrolysis rate constants and interferometric points for organic compounds.
|
| Benefits |
- More widely applicable models for predicting pollutant transport.
- A tool for better understanding and interpretation of process mechanisms and pathways.
|
Top
|
