Heart mediastinal and epicardial fat tissues are related to several adverse metabolic effects and cardiovascular risk factors, especially coronary artery disease (CAD). The manual segmentation of those fats is that the high dependence on user intervention and time-consuming analyzes. As a result, the automated measurement of cardiac fats could be considered as one of the most important biomarkers for cardiovascular risks in imaging and medical visualization by physicians. In this paper, we validate an automatic approach for the cardiac fat segmentation in non-contrast CT images then investigate the correlation between cardiac fat volume and CAD using the association rule mining algorithm. The pre-processing step includes threshold and contrast enhancement, the feature extraction step includes Gabor filter bank based on GLCM, the cardiac fat segmentation step is predicated on pattern recognition classification algorithms, and eventually, the step of investigating the relationship between cardiac fat volume and CAD is using FP-Growth algorithm. Experimental validation using CT images of two databases points to a good performance in cardiac fat segmentation. Experiments showed that the accuracy of the designed algorithm using the ensemble classifier with the best performance over other classifiers for the cardiac fat segmentation was 99.2%, with a sensitivity of 96.3% and a specificity of 99.8%. The results of using the FP-Growth algorithm showed that the low volume of epicardial (Confidence = 0.6818, Lift = 1.0626) and mediastinal (Confidence = 0.6696, Lift = 1.0436) fat are associated with healthy individuals and the high volume of epicardial (Confidence = 0.8, Lift = 2.2326) and mediastinal (Confidence = 0.75, Lift = 2.093) fat are related to individuals of CAD. As a result, cardiac fats can be used as a reliable biomarker tool in predicting the extent of CAD stenosis.

Medical image segmentation is an essential step in various diagnostic and treatment procedures. This study aimed to conduct a systematic review of state-of-the-art segmentation methods based on the target. The target complexity is a considerable challenge in medical image segmentation and the first issue that experts confront in diagnosing or treating patients. Additionally, each group of targets has similar characteristics, motivating to provide a target-based review to compare the deep-learning (DL)-based studies. This is the first time that a target-based review of medical image segmentation has provided a focus on recent DL developments. This study categorized publications into three targets: tumors, vessels, and pathological. Using a PRISMA strategy while considering the inclusion and exclusion criteria, 118 articles were identified on Google Scholar and PubMed from 2015 to 2023 in the fields of brain, liver, and lung tumors, blood vessels, and pathology image segmentation. This review could assist researchers in selecting the proper network and being aware of possible challenges. We also concluded that medical image segmentation using DL as a cross-disciplinary field is involved with both complex medical data and technical issues. Consequently, new interpretable approaches may be able to bridge the gap between medical specialists and artificial intelligence researchers.

Tibial plateau fractures account for approximately 1% of skeletal fractures, with treatment strategies varying based on fracture type, displacement, and articular involvement. Diagnosis is labor-intensive, time-consuming, repetitive, and subject to considerable inter-observer variability. Automated and precise approaches could improve accuracy and efficiency in fracture severity classification. With advances in artificial intelligence (AI), especially deep learning, such techniques are increasingly applied in medicine, yet their performance depends on high-quality training data. Here, we present a first-of-its-kind open-access dataset for AI-based analysis of tibial plateau fractures. The dataset comprises 421 heterogeneous anterior-posterior radiographs from 186 patients (mean age 45.88 ± 17.54 years; 37 females, 149 males), including normal and fractured knees. Fractures were classified by expert orthopedic surgeons and radiologists using the Schatzker system: type I (14.51%), II (18.27%), III (6.45%), IV (5.91%), V (6.45%), VI (17.20%), and normal (31.18%). All images were segmented to generate tibial bone masks, supporting morphological analysis, AI training, and automated fracture assessment. This dataset facilitates AI-driven fracture detection, classification, preoperative planning, and orthopedic assistant education.

Epicardial and Pericardial visceral adipose tissues around the cardiac are associated with several cardiovascular risk factors and adverse metabolic effects. As a result, automatic recognition and determination of cardiac fats can be used as an additional feature in medical imaging and visualization by physicians, due to advanced medical imaging technologies that offer a significant amount of information with high accuracy. The determination of cardiac fats in CT images also requires image segmentation methods. In this paper, we propose an automated method for the determination and segmentation of epicardial and pericardial adipose tissues in non-contrast cardiac CT images. The proposed method in this work includes the pre-processing step using thresholding in the fat range and contrast enhancement using histogram analysis, feature extraction step based on texture features extracted using Gabor filter bank based GLCM and pixel information, and the cardiac fats segmentation step are based on pixel labeling and classification algorithms. The experimental results also indicate a good performance of cardiac fat segmentation compared to the manual segmentation obtained by Expert. Experiments showed that the average accuracy obtained for epicardial and pericardial fats was 99% with a mean true positive rate of 98.6%. On average, the Dice Similarity Index was 98.8%.