Deep Convolutional Neural Networks for left ventricle segmentation.

Left ventricle (LV) segmentation is crucial for quantitative cardiac function analysis. Manual segmentation of the endocardium and epicardium is highly cumbersome; physicians limit delineation to the end-diastolic and end-systolic phases. A fully automated system could provide an analysis of cardiac morphology for all phases in a much shorter time. Most of the current LV segmentation methods are semi-automated and require error prone manual initialization. A fully-automated LV segmentation method would expedite the functional analysis of the LV, reduce subjectivity and improve patient experience. We automatically segment the LV wall in cardiac MRI images with a Deep Convolutional Neural Network (DCNN). This algorithm first calculates the probability of a pixel belonging to the LV wall or background and then generates a label based on those probabilities without manual initialization. We then compare these results to the results obtained with another DCNN initialization method using Gabor filters. With Gabor DCNN we obtain an accuracy of 0.97, specificity of 0.984, sensitivity of 0.841 and mean accuracy of 0.902. This shows that Gabor filters perform better than random filters in the DCNN for LV segmentation.

Vessel extraction in X-ray angiograms using deep learning.

Coronary artery disease (CAD) is the most common type of heart disease which is the leading cause of death all over the world. X-ray angiography is currently the gold standard imaging technique for CAD diagnosis. These images usually suffer from low quality and presence of noise. Therefore, vessel enhancement and vessel segmentation play important roles in CAD diagnosis. In this paper a deep learning approach using convolutional neural networks (CNN) is proposed for detecting vessel regions in angiography images. Initially, an input angiogram is preprocessed to enhance its contrast. Afterward, the image is evaluated using patches of pixels and the network determines the vessel and background regions. A set of 1,040,000 patches is used in order to train the deep CNN. Experimental results on angiography images of a dataset show that our proposed method has a superior performance in extraction of vessel regions.

Robust catheter identification and tracking in X-ray angiographic sequences.

Coronary artery disease (CAD) is one of the major causes of death worldwide. Today X-ray angiography is a standard method for CAD diagnosis. Usually, the quality of these images is not good enough. Noise, camera and heart motions, non-uniform illumination and even the presence of catheter are sources of quality degradation. The existence of catheter can produce difficulties in vessel extraction methods because catheter is structurally similar to arteries. In this paper we propose a fully automatic method for catheter detection and tracking during the whole angiography sequence. In this method with a vesselness map, we smooth each frame using guided filter. The catheter is detected in the first frame using Hough transform. We then fit a second order polynomial on the catheter and accurately track it throughout the sequence. Our method is tested on 25 X-ray angiography sequences where a precision of 0.9597 is achieved.

Relationship of treatment delays and mortality in patients undergoing fibrinolysis and primary percutaneous coronary intervention. The Global Registry of Acute Coronary Events.

OBJECTIVE: Treatment delays may result in different clinical outcomes in patients with ST-segment elevation myocardial infarction (STEMI) who receive fibrinolytic therapy vs primary percutaneous coronary intervention (PCI). The aim of this analysis was to examine how treatment delays relate to 6-month mortality in reperfusion-treated patients enrolled in the Global Registry of Acute Coronary Events (GRACE). DESIGN: Prospective, observational cohort study. SETTING: 106 hospitals in 14 countries. PATIENTS: 3959 patients who presented with STEMI within 6 h of symptom onset and received reperfusion with either a fibrin-specific fibrinolytic drug or primary PCI. MAIN OUTCOME MEASURES: 6-month mortality. METHODS: Multivariable logistic regression was used to assess the relationship between outcomes and treatment delay separately in each cohort, with time modelled with a quadratic term after adjusting for covariates from the GRACE risk score. RESULTS: A total of 1786 (45.1%) patients received fibrinolytic therapy, and 2173 (54.9%) underwent primary PCI. After multivariable adjustment, longer treatment delays were associated with a higher 6-month mortality in both fibrinolytic therapy and primary PCI patients (p<0.001 for both cohorts). For patients who received fibrinolytic therapy, 6-month mortality increased by 0.30% per 10-min delay in door-to-needle time between 30 and 60 min compared with 0.18% per 10-min delay in door-to-balloon time between 90 and 150 min for patients undergoing primary PCI. CONCLUSIONS: Treatment delays in reperfusion therapy are associated with higher 6-month mortality, but this relationship may be even more critical in patients receiving fibrinolytic therapy.

The role of hospital volume in coronary artery bypass grafting: is more always better?

OBJECTIVES: The goal of this study was to determine whether outcomes of nonemergent coronary artery bypass grafting (CABG) differed between low- and high-volume hospitals in patients at different levels of surgical risk. BACKGROUND: Regionalizing all CABG surgeries from low- to high-volume hospitals could improve surgical outcomes but reduce patient access and choice. "Targeted" regionalization could be a reasonable alternative, however, if subgroups of patients that would clearly benefit from care at high-volume hospitals could be identified. METHODS: We assessed outcomes of CABG at 56 U.S. hospitals using 1997 administrative and clinical data from Solucient EXPLORE, a national outcomes benchmarking database. Predicted in-hospital mortality rates for subjects were calculated using a logistic regression model, and subjects were classified into five groups based on surgical risk: minimal (< 0.5%), low (0.5% to 2%), moderate (2% to 5%), high (5% to 20%), and severe (> or =20%). We assessed differences in in-hospital mortality, hospital costs and length of stay between low- and high-volume facilities (defined as > or =200 annual cases) in each of the five risk groups. RESULTS: A total of 2,029 subjects who underwent CABG at 25 low-volume hospitals and 11,615 subjects who underwent CABG at 31 high-volume hospitals were identified. Significant differences in in-hospital mortality were seen between low- and high-volume facilities in subjects at moderate (5.3% vs. 2.2%; p = 0.007) and high risk (22.6% vs. 11.9%; p = 0.0026) but not in those at minimal, low or severe risk. Hospital costs and lengths of stay were similar across each of the five risk groups. Based on these results, targeted regionalization of subjects at moderate risk or higher to high-volume hospitals would have resulted in an estimated 370 transfers and avoided 16 deaths; in contrast, full regionalization would have led to 2,029 transfers and avoided 20 deaths. CONCLUSIONS: Targeted regionalization might be a feasible strategy for balancing the clinical benefits of regionalization with patients' desires for choice and access.

Coronary artery bypass grafting in Native Americans: a higher risk of death compared to other ethnic groups?

BACKGROUND: While the efficacy and safety of coronary artery bypass grafting (CABG) has been established in several clinical trials, little is known about its outcomes in Native Americans. MEASUREMENTS AND MAIN RESULTS: We assessed clinical outcomes associated with CABG in 155 Native Americans using a national database of 18,061 patients from 25 nongovernmental, not-for-profit U.S. health care facilities. Patients were classified into five groups: 1) Native American, 2) white, 3) African American, 4) Hispanic, and 5) Asian. We evaluated for ethnic differences in in-hospital mortality and length of stay, and after adjusting for age, gender, surgical priority, case-mix severity, insurance status, and facility characteristics (volume, location, and teaching status). Overall, we found the adjusted risk for in-hospital death to be higher in Native Americans when compared to whites (odds ratio [OR], 3.8; 95% confidence interval [CI], 1.5 to 9.8), African Americans (OR, 3.4; 95% CI, 1.1 to 9.9), Hispanics (OR, 7.1; 95% CI, 2.5 to 20.3), and Asians (OR, 2.8; 95% CI, 1.1 to 7.0). No significant differences were found in length of stay after adjustment across ethnic groups. CONCLUSIONS: The risk of in-hospital death following CABG may be higher in Native Americans than in other ethnic groups. Given the small number of Native Americans in the database (n = 155), however, further research will be needed to confirm these findings.

Case of the month.

Each month, we will present a challenging Case of the Month for Green Journal readers, who must use their clinical acumen to arrive at the correct answer. We will also post the case each month on the Journal's web site (http://www.elsevier.com/ajmselect). Several possible answers may be consistent with the case presentation; use your best judgment. Please send your answer (one per respondent) to The Green Journal at editors@amjmed.org or via FAX to (415) 447-2799. Indicate the case to which you are responding and include your complete address. The correct answer will appear in the next issue of the Journal. The first five persons who submit correct answers will receive a free one-year subscription to the Journal. Because of the volume of answers we receive, neither correct nor incorrect answers can be individually acknowledged. Colleagues of Drs. Green, Nallamothu, and Shea are not eligible for this month's case. If you would like to contribute a case, please submit a brief synopsis (<250 words) to the editorial office.

Electron-beam computed tomography in the diagnosis of coronary artery disease: a meta-analysis.

BACKGROUND: Electron-beam computed tomography (EBCT) is a new, noninvasive method of detecting coronary artery calcification that is being increasingly advocated as a diagnostic test for coronary artery disease (CAD). Before its clinical use is justified, however, the overall accuracy of EBCT must be better defined. OBJECTIVE: To estimate the accuracy of EBCT in diagnosing obstructive CAD. DATA SOURCES: English-language studies from January 1, 1979, through February 29, 2000, were retrieved using MEDLINE and Current Contents databases, bibliographies, and expert consultation. STUDY SELECTION: We included a study if it (1) used EBCT as a diagnostic test; (2) reported cases in absolute numbers of true-positive, false-positive, true-negative, and false-negative results; and (3) used coronary angiography as the reference standard for diagnosing obstructive CAD (defined as > or = 50% diameter stenosis). DATA EXTRACTION: Data were extracted from the included articles by 2 independent reviewers. DATA SYNTHESIS: Weighted pooled analysis and summary receiver operating characteristic (ROC) curve analysis were used to determine sensitivity and specificity rates. Results from 9 studies with 1662 subjects were included. Pooled sensitivity for EBCT was 92.3% (95% confidence interval [CI], 90.7%-94.0%) and pooled specificity was 51.2% (95% CI, 47.5%-54.9%). Maximum joint sensitivity and specificity for EBCT from its summary ROC curve was 75%. As the threshold for defining an abnormal test varied, sensitivity and specificity changed. For a threshold that resulted in a sensitivity of 90%, specificity was 54%; when sensitivity was 80%, specificity rose to 71%. CONCLUSION: The performance of EBCT as a diagnostic test for obstructive CAD is reasonable based on sensitivity and specificity rates from its summary ROC curve.

Potential clinical and economic effects of homocyst(e)ine lowering.

BACKGROUND: Elevated total homocyst(e)ine levels (>/=11 micromol/L) have been identified as a potential risk factor for coronary heart disease. However, the benefits expected from lowering homocyst(e)ine levels with folic acid and vitamin B(12) supplementation have yet to be demonstrated in clinical trials. SUBJECTS AND METHODS: We constructed a decision analytic model to estimate the clinical benefits and economic costs of 2 homocyst(e)ine-lowering strategies: (1) "treat all"-no screening, daily supplementation with folic acid (400 microg) and vitamin B(12) (cyanocobalamin; 500 microg) for all; (2) "screen and treat"-screening, followed by daily supplementation with folic acid and vitamin B(12) for individuals with elevated homocyst(e)ine levels. Simulated cohorts of 40-year-old men and 50-year-old women in the general population were evaluated. In the base-case analysis, we assumed that lowering elevated levels would reduce excess coronary heart disease risk by 40%; however, this assumption and others were evaluated across a broad range of potential values using sensitivity analysis. Primary outcomes were discounted costs per life-year saved. RESULTS: Although the treat-all strategy was slightly more effective overall, the screen and treat strategy resulted in a much lower cost per life-year saved ($13,600 in men and $27,500 in women) when compared with no intervention. Incremental cost-effectiveness ratios for the treat-all strategy compared with the screen and treat strategy were more than $500,000 per life-year saved in both cohorts. Sensitivity analysis showed that cost-effectiveness ratios for the screen and treat strategy remained less than $50,000 per life-year saved under several unfavorable scenarios, such as when effective homocyst(e)ine lowering was assumed to reduce the relative risk of coronary heart disease-related death by only 11% in men or 23% in women. CONCLUSIONS: Homocyst(e)ine lowering with folic acid and vitamin B(12) supplementation could result in substantial clinical benefits at reasonable costs. If homocyst-(e)ine lowering is considered, a screen and treat strategy is likely to be more cost-effective than universal supplementation. Arch Intern Med. 2000;160:3406-3412.

Heterotopic bone formation in male and female rabbits.

We performed mechanical induction of heterotopic ossification in the quadriceps of the right hind limb of six male and six female sexually mature New Zealand white rabbits of similar size. The effect of sex difference on heterotopic bone was assessed by analyzing plain radiographs of the femur. The results indicate that a male/female sex difference in heterotopic bone formation cannot be demonstrated.

Effect of two nonsteroidal antiinflammatory drugs on heterotopic bone formation in a rabbit model.

Using the model described by Michelsson, 24 mature New Zealand white rabbits underwent mechanical induction of heterotopic ossification in the quadriceps of the right hind limb. The rabbits were equally divided into four groups: indomethacin-treated, piroxicam-treated, and their respective controls. The effect of drug therapy on the production of heterotopic bone was assessed by analyzing plain radiographs of the femur using the grading system described by Scott. The results demonstrated that, while indomethacin is effective in decreasing the formation of heterotopic bone, piroxicam--when used in the dosage previously demonstrated to have no deleterious effect on healing bone--is not. The analysis of piroxicam blood levels demonstrated that the experimental piroxicam dose is inadequate to produce effective plasma blood levels in the rabbit. Therefore, any potential benefit from using piroxicam, as opposed to other nonsteroidal antiinflammatory drugs regarding fracture healing and bone remodeling, cannot be applied for the prophylactic treatment of heterotopic bone formation.

Hypothermia following acute carbon-monoxide poisoning increases mortality.

The effect of body temperature during recovery from acute severe carbon monoxide (CO) poisoning on morbidity and mortality was investigated using an unanesthetized animal model. Levine prepared female rats were exposed to 2700 ppm CO for 90 min, displaying the usual hypothermia, hypotension and bradycardia. Body temperature in survivors was either maintained at the terminal CO exposure value for 4 h ('cooled'), rapidly raised to the pre-CO exposure value for 4 h ('heated') or allowed to return to normal naturally ('unheated'). Following this period of recovery, the 'heated' rats sustained the lowest mortality rate (22%), the 'unheated' rats a greater mortality rate (44%), and the 'cooled' rats the highest mortality rate (50%). Inclusive of the surviving rats, there were no significant differences in neurologic deficit between the three groups, after 4, 24 or 48 h of recovery. Aside from a significant delay in recovery of blood pressure in the 'heated' rats relative to the other two treatment groups post-CO, there were no differences in blood pressure, heart rate or plasma glucose. The results suggest that rapid return to euthermia improves survival following acute severe CO poisoning, contrasting with an earlier study showing the detrimental effect of euthermia during CO poisoning.