RESUMO
Background: Inflammation and lipid infiltration play crucial roles in the development of atherosclerosis. This study aimed to investigate the association between various complex indexes of blood cell types and lipid levels with the severity of coronary artery stenosis and their predictive value in coronary heart disease (CHD). Methods: The retrospective study was conducted on 3,201 patients who underwent coronary angiography at the Department of Zhongnan Hospital of Wuhan University. The patients were divided into two groups: CHD group and non-CHD group. The CHD group was further classified into three subgroups (mild, moderate, severe) based on the tertiles of their Gensini score or SYNTAX score I. Various complex indexes of blood cell types and lipid levels were compared between the groups. Results: It revealed a positive correlation between all complex indexes and the severity of coronary artery stenosis. The systemic inflammation-response index/high-density lipoprotein cholesterol count (SIRI/HDL) exhibited the strongest correlation with both severity scores (Gensini score: r = 0.257, P < 0.001; SYNTAX score I: r = 0.171, P < 0.001). The monocyte to high-density lipoprotein cholesterol ratio (MHR) was identified as a stronger independent risk factor for CHD. However, SIRI/HDL had higher diagnostic efficacy for CHD (sensitivity 66.7%, specificity 60.4%, area under curve 0.680, 95% CI: 0.658-0.701). Notably, the pan-immune-inflammation value multiplied by low-density lipoprotein cholesterol count (PIV × LDL) exhibited the highest sensitivity of 85.2%. Conclusion: All complex indexes which we investigated exhibited positive correlations with the severity of coronary artery stenosis. SIRI/HDL demonstrated higher diagnostic efficiency for CHD and a significant correlation with the severity of coronary artery stenosis.
RESUMO
Purpose: This study implements and demonstrates a deep learning (DL) approach for screening referable horizontal strabismus based on primary gaze photographs using clinical assessments as a reference. The purpose of this study was to develop and evaluate deep learning algorithms that screen referable horizontal strabismus in children's primary gaze photographs. Methods: DL algorithms were developed and trained using primary gaze photographs from two tertiary hospitals of children with primary horizontal strabismus who underwent surgery as well as orthotropic children who underwent routine refractive tests. A total of 7026 images (3829 non-strabismus from 3021 orthoptics [healthy] subjects and 3197 strabismus images from 2772 subjects) were used to develop the DL algorithms. The DL model was evaluated by 5-fold cross-validation and tested on an independent validation data set of 277 images. The diagnostic performance of the DL algorithm was assessed by calculating the accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC). Results: Using 5-fold cross-validation during training, the average AUCs of the DL models were approximately 0.99. In the external validation data set, the DL algorithm achieved an AUC of 0.99 with a sensitivity of 94.0% and a specificity of 99.3%. The DL algorithm's performance (with an accuracy of 0.95) in diagnosing referable horizontal strabismus was better than that of the resident ophthalmologists (with accuracy ranging from 0.81 to 0.85). Conclusions: We developed and evaluated a DL model to automatically identify referable horizontal strabismus using primary gaze photographs. The diagnostic performance of the DL model is comparable to or better than that of ophthalmologists. Translational Relevance: DL methods that automate the detection of referable horizontal strabismus can facilitate clinical assessment and screening for children at risk of strabismus.
