The use of physician domain knowledge to improve the learning of rule-based models for decision-support.

This paper describes a study testing the hypothesis that the learning of a decision-support model by a computer learning algorithm from clinical data can be improved by the addition of domain knowledge from practicing physicians. The domain of the experiment is community-acquired pneumonia. The overall design of the study compares a computer learning algorithm given clinical data to one given clinical data plus domain knowledge added by physician subjects. This study showed that the performance of the computer-generated models augmented with knowledge added by physician subjects were significantly better than the computer-generated models generated without added knowledge using a two-stage rule induction algorithm in the domain of community-acquired pneumonia. This result was highly significant and shows that the addition of domain knowledge may be beneficial to the learning of clinical decision-support models, especially in domains where data is limited.

An evaluation of machine-learning methods for predicting pneumonia mortality.

This paper describes the application of eight statistical and machine-learning methods to derive computer models for predicting mortality of hospital patients with pneumonia from their findings at initial presentation. The eight models were each constructed based on 9847 patient cases and they were each evaluated on 4352 additional cases. The primary evaluation metric was the error in predicted survival as a function of the fraction of patients predicted to survive. This metric is useful in assessing a model's potential to assist a clinician in deciding whether to treat a given patient in the hospital or at home. We examined the error rates of the models when predicting that a given fraction of patients will survive. We examined survival fractions between 0.1 and 0.6. Over this range, each model's predictive error rate was within 1% of the error rate of every other model. When predicting that approximately 30% of the patients will survive, all the models have an error rate of less than 1.5%. The models are distinguished more by the number of variables and parameters that they contain than by their error rates; these differences suggest which models may be the most amenable to future implementation as paper-based guidelines.

The use of misclassification costs to learn rule-based decision support models for cost-effective hospital admission strategies.

Cost-effective health care is at the forefront of today's important health-related issues. A research team at the University of Pittsburgh has been interested in lowering the cost of medical care by attempting to define a subset of patients with community-acquire pneumonia for whom outpatient therapy is appropriate and safe. Sensitivity and specificity requirements for this domain make it difficult to use rule-based learning algorithms with standard measures of performance based on accuracy. This paper describes the use of misclassification costs to assist a rule-based machine-learning program in deriving a decision-support aid for choosing outpatient therapy for patients with community-acquired pneumonia.

A calorimetric approach to the study of the interactions of cytidine-3'-phosphate with bovine seminal ribonuclease.

A calorimetric study at 25 degrees C is reported on the interaction between allosteric bovine seminal ribonuclease and cytidine-3'-phosphate. The results are compared with those obtained under identical experimental conditions for the interaction of pancreatic ribonuclease A and the same nucleotide. The analysis of the data provides evidence that the binding sites of seminal ribonuclease for cytidine-3'-phosphate are not equivalent, in agreement with previous equilibrium dialysis studies. A model with two sites with different affinities toward the nucleotide, the site with higher affinity resembling the binding site of pancreatic ribonuclease, is proposed. The values calculated for the thermodynamic parameters provide an insight of the forces involved in the interaction of the two enzymes with the nucleotide.

Protein-ligand interaction. A calorimetric study of the interaction of oligosaccharides and hen ovalbumin glycopeptides with concanavalin A.

A calorimetric study is reported concerning the interaction between concanavalin A (Con A) and some oligosaccharides and glycopeptides hydrolyzed from hen ovalbumin. The measurements were carried out in acetate buffer, pH 4.5, where, by far, the prevailing form of the protein is the dimeric one [Kalb, A.J., & Lustig, A. (1968) Biochim. Biophys. Acta 168, 366; Dani, M., Manca, F., & Rialdi, G. (1981) Biochim. Biophys. Acta 667, 108]. The calorimetric technique allows the direct determination of the binding enthalpy delta H, degrees B, the evaluation of the apparent association constant K'B, and then the evaluation of the apparent free energy and entropy, delta G degrees' B and delta S degrees' B. Three groups of data have been collected in the present study. The first one concerns the interaction between concanavalin A and some mono- and disaccharides [methyl alpha-glucopyranoside (alpha MGlup), methyl alpha-mannopyranoside (alpha MManp), D-maltose, D-trehalose, and D-cellobiose]. The analysis of the data indicates that in these cases there are small favorable entropic and enthalpic contributions to the affinity. The stoichiometry of the reaction is 2 mol of ligand/mol of Con A dimer, the sites resulting being equivalent and noninteracting. Melezitose, the only trisaccharide studied, shows a different behavior: its affinity for Con A is higher as compared to the other oligosaccharides containing alpha-glucosyl residues and closer to that of methyl alpha-mannopyranoside. However, the stoichiometry is different, namely, 1 mol of ligand/dimer of Con A.(ABSTRACT TRUNCATED AT 250 WORDS)