Deep learning to detect significant coronary artery disease from plain chest radiographs

Background

Coronary artery disease (CAD) is the single leading cause of mortality, premature death, and morbidity worldwide. Artificial intelligence (AI) could help identify markers present within first-line diagnostic imaging routinely performed in patients referred for suspected angina, such as chest radiographs.

Purpose

To train, test, and validate a deep learning (DL) algorithm for detecting the presence of significant CAD based on chest radiographs.

Methods

Data of patients undergoing chest radiography and coronary angiography were retrospectively analysed. A deep convolutional neural network (DCNN) was designed to detect significant CAD from the patient posteroanterior/anteroposterior chest radiograph. The DCNN was trained for binary classification of severe CAD absence/presence (at least one diseased coronary vessel with ≥70% stenosis). Coronary angiography reports were used as the ground truth. Sensitivity, specificity, and area under the receiver operating characteristic curve (AUC) of the DCNN were calculated. Multivariate analysis was performed to identify independent correlation among the presence of significant CAD (dependent variable), DCNN prediction, and CAD risk factors.

Results

Information of 7728 patients referred for suspected angina was reviewed. Severe CAD was present in 4482 patients (58%; 1% left main, 28% one vessel, 16% two vessels, and 12% 3 vessels). Patients were randomly divided for training (70%; n=5454) and fine-tuning/testing (10%; n=773) of the algorithm. Internal validation was performed with the remaining patients (20%; n=1501). At binary logistic regression, the DCNN prediction was the strongest independent determinant of severe CAD (p<0.0001; OR: 52.8; CI: 25.1–110.9). Age (p=0.008; OR: 1.01; CI: 1.0–1.02) and Diamond-Forrester score (p<0.0001; OR: 1.022; CI: 1.018–1.026) were also independently related to CAD, although with lower significance and odds-ratios. Using an operating cut-point with high sensitivity, the DCNN had a sensitivity of 0.90 and specificity of 0.31 to detect significant CAD in the internal validation group (AUC 0.73; 95% CI DeLong, 0.69–0.76). Adding to the AI chest radiograph interpretation, patient age and angina status improved the prediction (AUC 0.77; 95% CI DeLong, 0.74–0.80).

Conclusion

The chest radiograph is ubiquitous and carries a plethora of information concerning the patient's health status, including direct and indirect signs of CAD. Our DL algorithm can predict, with high sensitivity, the presence of severe CAD in patients referred for suspected angina. It could be used to pre-test significant CAD probability in outpatient clinics, emergency room settings, and CAD screening in more extensive settings. Further studies are required to externally validate the algorithm and develop a clinically applicable tool.

Funding Acknowledgement

Type of funding sources: None.

Completely isolated enteric duplications cysts: a survey of four cases

Enteric duplication cysts are uncommon congenital abnormalities with epithelial lining. They are cystic or tubular structures intimately attached to a portion of the gastrointestinal tract; they are usually located on the mesenteric site of the digestive tract sharing common blood supply. Isolated cystic duplications are an extremely rare variant with their own blood supply: in literature only five cases have been reported. We present our four cases series of this uncommon anomaly.

Lymphatic preservation using methylene blue dye during varicocele surgery: a single-center retrospective study

OBJECTIVE

Hydrocele and testicular edema caused by division of lymphatic vessels during varicocelectomy could lead to decrease in testicular function. In-vivo methylene blue mapping of testicular lymphatic vessels should prevent damage to the lymphatic system.

MATERIALS AND METHODS

We retrospectively compared outcomes for 46 patients who received an intraparenchymal injection of 0.25 ml of vital dye (isosulphan blue) before a laparoscopic or an inguinal/subinguinal spermatic vein ligation with 93 controls in whom no mapping technique was adopted.

RESULTS

Methylene blue mapping of testicular lymphatics reduced the incidence of postvaricocelectomy hydrocele from 6.4% (6/93) to 2.1% (1/46); the incidence of hydrocele was 0% in all cases of successful lymphatic mapping.

CONCLUSION

Mapping of testicular lymphatic drainage with intraparenchymal vital dye is an easy, safe, rapid and cost-free technique. We stress the importance of sparing the lymphatic system to ensure the best andrological outcome.

Cell cycle- and cell growth-regulated proteolysis of mammalian CDC6 is dependent on APC-CDH1

CDC6 is conserved during evolution and is essential and limiting for the initiation of eukaryotic DNA replication. Human CDC6 activity is regulated by periodic transcription and CDK-regulated subcellular localization. Here, we show that, in addition to being absent from nonproliferating cells, CDC6 is targeted for ubiquitin-mediated proteolysis by the anaphase promoting complex (APC)/cyclosome in G(1). A combination of point mutations in the destruction box and KEN-box motifs in CDC6 stabilizes the protein in G(1) and in quiescent cells. Furthermore, APC, in association with CDH1, ubiquitinates CDC6 in vitro, and both APC and CDH1 are required and limiting for CDC6 proteolysis in vivo. Although a stable mutant of CDC6 is biologically active, overexpression of this mutant or wild-type CDC6 is not sufficient to induce multiple rounds of DNA replication in the same cell cycle. The APC-CDH1-dependent proteolysis of CDC6 in early G(1) and in quiescent cells suggests that this process is part of a mechanism that ensures the timely licensing of replication origins during G(1).

Genetic aspects of the worldwide colonization process of Ceratitis capitata

Multilocus enzyme electrophoresis data from 26 polymorphic loci (124 alleles) were used to analyze the genetic aspects of the worldwide colonization of Ceratitis capitata (medfly). Eighty-two samples of 17 populations were collected from six regions throughout the species range: Africa, extra-Mediterranean islands (Madeira and Gran Canaria), Mediterranean region, Latin America (Guatemala), Pacific (Hawaii), and Australia. The variability parameters (H, P, A) reveal that the geographical dispersal of medfly from its ancestral source area (East Africa) is associated with a great reduction in variability. The pattern of decreasing variability occurs at two regional levels: in the African-Mediterranean region where the differentiation is gradual, and in the Latin American-Pacific region where some ancestral variability is still present as a consequence of recent colonization. The UPGMA phylogenetic tree, derived from Nei's genetic distances, shows the presence of intraspecific differentiative processes affecting mainly the two island populations, Réunion and Hawaii. The population genetic changes observed in the species range are consistent with both the chronology and the historical circuitous course of the medfly colonization process.