Evaluation of a breast cancer computer aided diagnosis system.

Mammography is accepted as the most effective method to detect breast cancer. However, interpreting a mammogram is not easy for not experienced radiologists. The aim of computer aided detection techniques in breast cancer is to improve the chance that a malignant region is detected and appropriately evaluated. Breast microcalcifications have been considered as a very useful index of malignancy, which helps in the early detection of breast cancer. A system of computer aided diagnosis has been developed that is based on detailed analysis and evaluation of related features of individual microcalcifications and of formed clusters helping the doctor to make risk estimation for each microcalcification cluster as well as for isolated microcalcifications. This information is considered to be very useful to radiologists, giving them extra input before making their estimation of each case. The aforementioned system has been thoroughly tested using a number of real life cases provided from collaborating doctors. Each case, apart from the mammograms, was accompanied by a biopsy test result, the patient's demographic data and medical history. A total of 200 cases (147 benign and 53 malignant) have been examined and the results are presented as the Receiver Operating Characteristic (ROC) performance and are quantified using the ROC curve. The system is showing high levels of sensitivity identifying correctly all malignant cases.

Effects of exercise rehabilitation program on heart rate recovery in patients with chronic heart failure.

BACKGROUND: Heart rate recovery (HRR1) immediately after exercise reflects parasympathetic activity, which is markedly attenuated in chronic heart failure (CHF) patients. The aim of our study was to examine both continuous and interval exercise training effects on HRR1 in these patients. DESIGN: The population study consisted of 29 stable CHF patients that participated at a rehabilitation program of 36 sessions, three times per week. Of the 29 patients, 24 completed the program. Patients were randomly assigned to interval {n=10 [100% peak work rate (WRp) for 30 s, alternating with rest for 30 s]} and to continuous training [n=14 (50%WRp)]. METHODS: All patients performed a symptom-limited cardiopulmonary exercise test on a cycle ergometer before and after the completion of the program. Measurements included peak oxygen uptake (VO2p), anaerobic threshold (AT), WRp, first degree slope of VO2 during the first minute of recovery (VO2/t-slope), chronotropic response [% chronotropic reserve (CR)=(peak HR - resting HR)x100/(220 - age - resting HR)], HRR1 (HR difference from peak exercise to one minute after). RESULTS: After the completion of the rehabilitation program there was a significant increase of WRp, VO2p, AT and VO2/t-slope (by 30%, P=0.01; 6%, P=0.01; 10%, P=0.02; and 27%, P=0.03 respectively for continuous training and by 21%, P<0.05; 8%, P=0.01; 6%, P=NS; and 48%, P=0.02 respectively for interval training). However, only patients exercised under the continuous training regime had a significant increase in HRR1 (15.0+/-9.0 to 24.0+/-12 bpm; P=0.02) and CR (57+/-19 to 72+/-21%, P=0.02), in contrast with those assigned to interval training (HRR1: 21+/-11 to 21+/-8 bpm; P=NS and CR: 57+/-18 to 59+/-21%, P=NS). CONCLUSIONS: Both continuous and interval exercise training program improves exercise capacity in CHF patients. However, continuous rather than interval exercise training improves early HRR1, a marker of parasympathetic activity, suggesting a greater contribution to the autonomic nervous system.

Early heart rate recovery after exercise predicts mortality in patients with chronic heart failure.

BACKGROUND: Patients with chronic heart failure (CHF) have multiple abnormalities of autonomic regulation that have been associated to their high mortality rate. Heart rate recovery immediately after exercise is an index of parasympathetic activity, but its prognostic role in CHF patients has not been determined yet. METHODS: Ninety-two stable CHF patients (83M/9F, mean age: 51+/-12 years) performed an incremental symptom-limited cardiopulmonary exercise testing. Measurements included peak O2 uptake (VO2p), ventilatory response to exercise (VE/VCO2 slope), the first-degree slope of VO2 for the 1st minute of recovery (VO2/t-slope), heart rate recovery [(HRR1, bpm): HR difference from peak to 1 min after exercise] and chronotropic response to exercise [%chronotropic reserve (CR, %)=(peak HR-resting HR/220-age-resting HR)x100]. Left ventricular ejection fraction (LVEF, %) was also measured by radionuclide ventriculography. RESULTS: Fatal events occurred in 24 patients (26%) during 21+/-6 months of follow-up. HRR1 was lower in non-survivors (11.4+/-6.4 vs. 20.4+/-8.1; p<0.001). All cause-mortality rate was 65% in patients with HRR112 bpm (log-rank: 32.6; p<0.001). By multivariate survival analysis, HRR1 resulted as an independent predictor of mortality (chi2=19.2; odds ratio: 0.87; p<0.001) after adjustment for LVEF, VO2p, VE/VCO2 slope, CR and VO2/t-slope. In a subgroup of patients with intermediate exercise capacity (VO2p: 10-18, ml/kg/min), HRR1 was a strong predictor of mortality (chi2: 14.3; odds ratio: 0.8; p<0.001). CONCLUSIONS: Early heart rate recovery is an independent prognostic risk indicator in CHF patients and could be used in CHF risk stratification.

VE/VCO2 slope is associated with abnormal resting haemodynamics and is a predictor of long-term survival in chronic heart failure.

BACKGROUND: Patients with chronic heart failure (CHF) present with exercise-induced hyperpnea, but its pathophysiological mechanism has not been thoroughly investigated. We aimed to determine the relationship between exercise-induced hyperpnea, resting haemodynamic measurements and the validity of ventilatory response (V(E)/V(CO(2)) slope) as a mortality predictor in CHF patients. METHODS: Ninety-eight CHF patients (90M/8F) underwent a symptom-limited treadmill cardiopulmonary exercise test (CPET). Right heart catheterization and radionuclide ventriculography were performed within 72 h of CPET. RESULTS: Twenty-seven patients died from cardiac causes during 20+/-6 months follow-up. Non-survivors had a lower peak oxygen consumption (V(O(2)p)), (16.5+/-4.9 vs. 20.2+/-6.1, ml/kg/min, p=0.003), a steeper V(E)/V(CO(2)) slope (34.8+/-8.3 vs. 28.9+/-4.8, p<0.001) and a higher pulmonary capillary wedge pressure (PCWP) (19.5+/-8.6 vs. 11.7+/-6.5 mm Hg, p=0.008) than survivors. By multivariate survival analysis, the V(E)/V(CO(2)) slope as a continuous variable was an independent prognostic factor (chi(2): 8.5, relative risk: 1.1, 95% CI: 1.03-1.18, p=0.004). Overall mortality was 52% in patients with V(E)/V(CO(2)) slope > or =34 and 18% in those with V(E)/V(CO(2)) slope <34 (log rank: 18.5, p<0.001). In a subgroup of patients (V(O(2)p): 10-18 ml/kg/min), V(E)/V(CO(2)) slope was a significant predictor of mortality (relative risk: 6.2, 95% CI: 1.7-22.2, p=0.002). Patients with high V(E)/V(CO(2)) slope had higher resting PCWP (19.9+/-9.1 vs. 11.3+/-5.7 mmHg, p<0.001) and V(E)/V(CO(2)) slope correlated significantly with PCWP (r: 0.57, p<0.001). CONCLUSIONS: The V(E)/V(CO(2)) slope, as an index of ventilatory response to exercise, improves the risk stratification of CHF patients. Interstitial pulmonary oedema may be a pathophysiological mechanism of inefficient ventilation during exercise in these patients.