Biomedical subjects

Peter J Fos

Publications and source records attributed to Peter J Fos.

3 recordsLinked to original sources

PubMed2003

Combining the performance strengths of the logistic regression and neural network models: a medical outcomes approach.

The assessment of medical outcomes is important in the effort to contain costs, streamline patient management, and codify medical practices. As such, it is necessary to develop predictive models that will make accurate predictions of these outcomes. The neural network methodology has often been shown to perform as well, if not better, than the logistic regression methodology in terms of sample predictive performance. However, the logistic regression method is capable of providing an explanation regarding the relationship(s) between variables. This explanation is often crucial to understanding the clinical underpinnings of the disease process. Given the respective strengths of the methodologies in question, the combined use of a statistical (i.e., logistic regression) and machine learning (i.e., neural network) technology in the classification of medical outcomes is warranted under appropriate conditions. The study discusses these conditions and describes an approach for combining the strengths of the models.

Artificial Intelligence↗

PubMed2003

Health-related quality of life of cataract patients: cross-cultural comparisons of utility and psychometric measures.

BACKGROUND: This study was conducted to assess the presence and/or absence of cross-cultural differences or similarities between Korean and United States cataract patients. A systematic assessment was performed using utility and psychometric measures in the study population. RESEARCH DESIGN: A cross-sectional study design was used to examine the comparison of preoperative outcomes measures in cataract patients in Korea and the United States. Study subjects were selected using non-probabilistic methods and included 132 patients scheduled for cataract surgery in one eye. PARTICIPANTS: Subjects were adult cataract patients at Samsung and Kunyang General Hospital in Seoul, Korea, and Tulane University Hospital and Clinics in New Orleans, Louisiana. MEASUREMENTS: Preoperative utility was assessed using the verbal rating scale and standard reference gamble techniques. Current preoperative health status was assessed using the SF-36 and VF-14 surveys. Current preoperative Snellen visual acuity was used as a clinical measure of vision status. RESULTS: Korean patients were more likely to be younger (p = 0.001), less educated (p = 0.001), and to have worse Snellen visual acuity (p = 0.002) than United States patients. Multivariate analysis of variance (MANOVA) revealed that in contrast to Korean patients, United States patients were assessed to have higher scoring in general health, vitality, VF-14, and verbal rating for visual health. This higher scoring trend persisted after controlling for age, gender, education and Snellen visual acuity. The difference in health-related quality of life (HRQOL) between the two countries was quite clear, especially in the older age and highly educated group. CONCLUSIONS: Subjects in Korea and the United States were significantly different in quality of life, functional status and clinical outcomes. Subjects in the United States had more favorable health outcomes than those in Korea. These differences may be caused by multiple factors, including country-specific differences in economic status, health care system, cultural value system, and health policy. Cross-cultural differences should be considered when making international comparisons of quality of life.

Aged↗

PubMedDate not supplied

Data mining a diabetic data warehouse.

Diabetes is a major health problem in the United States. There is a long history of diabetic registries and databases with systematically collected patient information. We examine one such diabetic data warehouse, showing a method of applying data mining techniques, and some of the data issues, analysis problems, and results. The diabetic data warehouse is from a large integrated health care system in the New Orleans area with 30,383 diabetic patients. Methods for translating a complex relational database with time series and sequencing information to a flat file suitable for data mining are challenging. We discuss two variables in detail, a comorbidity index and the HgbA1c, a measure of glycemic control related to outcomes. We used the classification tree approach in Classification and Regression Trees (CART) with a binary target variable of HgbA1c >9.5 and 10 predictors: age, sex, emergency department visits, office visits, comorbidity index, dyslipidemia, hypertension, cardiovascular disease, retinopathy, end-stage renal disease. Unexpectedly, the most important variable associated with bad glycemic control is younger age, not the comorbiditity index or whether patients have related diseases. If we want to target diabetics with bad HgbA1c values, the odds of finding them is 3.2 times as high in those <65 years of age than those older. Data mining can discover novel associations that are useful to clinicians and administrators [corrected].

Adult↗