R. L. Lee

Long-term effects of glucosamine sulphate on osteoarthritis progression: a randomised, placebo-controlled clinical trial

BACKGROUND Treatment of osteoarthritis is usually limited to short-term symptom control. We assessed the effects of the specific drug glucosamine sulphate on the long-term progression of osteoarthritis joint structure changes and symptoms. METHODS We did a randomised, double-blind placebo controlled trial, in which 212 patients with knee osteoarthritis were randomly assigned 1500 mg sulphate oral glucosamine or placebo once daily for 3 years. Weightbearing, anteroposterior radiographs of each knee in full extension were taken at enrolment and after 1 and 3 years. Mean joint-space width of the medial compartment of the tibiofemoral joint was assessed by digital image analysis, whereas minimum joint-space width--ie, at the narrowest point--was measured by visual inspection with a magnifying lens. Symptoms were scored by the Western Ontario and McMaster Universities (WOMAC) osteoarthritis index. FINDINGS The 106 patients on placebo had a progressive joint-space narrowing, with a mean joint-space loss after 3 years of -0.31 mm (95% CI -0.48 to -0.13). There was no significant joint-space loss in the 106 patients on glucosamine sulphate: -0.06 mm (-0.22 to 0.09). Similar results were reported with minimum joint-space narrowing. As assessed by WOMAC scores, symptoms worsened slightly in patients on placebo compared with the improvement observed after treatment with glucosamine sulphate. There were no differences in safety or reasons for early withdrawal between the treatment and placebo groups. INTERPRETATION The long-term combined structure-modifying and symptom-modifying effects of gluosamine sulphate suggest that it could be a disease modifying agent in osteoarthritis.

Incorporating Image Processing in a Clinical Decision Support System

A prototype system to assist radiologists in the differential diagnosis of mammographic calcifications is presented. Our approach is to incorporate image-processing operators within a knowledge-based decision support system. The work described in this paper involves three stages. The first is to identify a set of terms that can represent the knowledge required in an example of radiological decision-making. The next is to identify image processing operators to extract the required information from the image. The final stage is to provide links between the set of symbolic terms and the image processing operators.

Incorporating Explanations into a CAD tool for the classification of calcifications

We argue that CAD systems should be designed as aids to decision-making. We have developed a prototype aid for the classification of calcifications that can present an explanation of the reasoning behind the classification. A study was carried out to identify a set of terms commonly used by radiologists to describe calcifications. A second study assessed the capacity of the these terms to discriminate between benign and malignant calcifications. A set of discriminating terms was identified. Image-processing measures were then implemented to identify calcifications for which the descriptors applied. Linear discriminant functions were derived for the classification of a set of 40 groups of calcifications. Two experiments were performed to test a prototype CAD system using the terms to advise on the differential diagnosis of calcfications. Image regions identifed by a commercial prompting tool were classified according to the descriptors assigned by the linear discriminant functions. A panel of expert radiologists reviewed the system and assessed its value as an aid to decision-making.

Incorporating radiological knowledge in a CAD system

Abstract This work aims to assist in the differential diagnosis of mammographic abnormalities. A prototype CAD system is being developed which will incorporate a knowledge base describing the characteristic features of benign and malignant calcifications. The knowledge base will provide advice to assist radiologists in assessing the significance of prompts placed by the CAD system. In this paper, we describe a study to elicit the radiological knowledge to be represented in the system. Using a set of descriptors derived in an earlier study, we obtained assessments and detailed descriptions of 40 sets of calcifications from 10 expert radiologists. This allowed us to identify a set of arguments relating descriptions of calcifications to assessments of the risk of malignancy. The study provides valuable data on the significance and salience of certain characteristics of mammographic calcifications. This data provides the basis of a mapping between image processing measurements and symbolic knowledge.