Evaluation of left ventricular mass measured by 3D echocardiography using magnetic resonance imaging as gold standard.

OBJECTIVE: Increased left ventricular mass (LVM) and presence of left ventricular hypertrophy (LVH) are predictors of cardiovascular morbidity and mortality, but can be reversed with proper treatment of the underlying cause. Therefore accurate as well as reproducible methods for diagnosis and follow-up are needed. We evaluated different modalities by which to measure LVM in patients with no known LVH using magnetic resonance imaging (MRI) as the gold standard: ECG using the formulae proposed by Sokolow-Lyon and Cornell, 2D echocardiography and 3D echocardiography. METHODS: 34 subjects were included in the study; 17 had a history of myocardial infarction, 7 had pulmonary hypertension and 10 were healthy. All patients and controls had a standard 12-lead ECG, a transthoracic 2D and 3D echocardiographic study and a cardiac MRI. RESULTS: ECG estimates of LVM correlated poorly with LVM by MRI (r = 0.18, NS and 0.16, NS for Sokolow-Lyon and Cornell, respectively), whereas a moderate correlation between 2D and 3D echocardiography and MRI was observed (r = 0.63, p<0.001 and r = 0.74, p<0.001, respectively). All methods were reproducible with no significant bias. CONCLUSION: LVM measured by 3D echocardiography is highly accurate compared to LVM measured by MRI. LVM calculated from 2D echocardiography also proved useful, whereas estimates of LVM by ECG are inaccurate in a non-hypertrophic population.

Techniques for rapid quantitative assessment of activity levels in small-group tutorials.

Two techniques for the rapid quantitative analysis of student participation in small-group teaching were investigated. In the first approach an observer, who also acted as a 'critical friend', recorded the length of individual contributions using a computer keyboard as a simple timing device. In the second approach, small-group sessions were recorded with a portable stereophonic audiotape recorder. The teacher was recorded on one channel, all students on the other. A computer program produced automated analysis of these small group interactions by computing relative amount of speech on each channel. Simple analysis produced automatically by the programs revealed the overall style of the tutorial--variably 'mini-lectures' by teachers with very little participation by the student body, rapid 'question and answer' sessions with about equal teacher/student body involvement or 'mini-presentations' by students with the teacher offering sparse comments in the manner of a facilitator. By presenting results in a graphic format, teachers can be given rapid objective feedback on their teaching style. Coupled with short verbal/non-verbal quizzes at the end of tutorials and information from other assessments, the value of using levels of participation as a measure of the efficiency of such small-group sessions can itself be assessed.

Language as a barrier to the acquisition of anatomical knowledge.

The language of instruction for medical students at the University of Hong Kong is English, which is their second language. This presents a potential barrier to their academic learning. We have studied the extent of this problem by looking at the amount of terminology students have to assimilate during their anatomy course and the way in which the terminology is structured and explained (this provides, more generally, an indication of the factual loading to which medical students are exposed). We have also investigated the effect of the quality of students' language skills on their examination results in anatomy class tests and MBBS exams. It was found that students' entrance levels in English correlated well with their final examination results and that the quality and quantity of their English also correlated highly with scores in class tests. On this basis, we conclude that language is one of the most important barriers to students' academic success in the department and that current teaching materials may not be appropriate.