An enhanced Swin Transformer for soccer player reidentification.

The re-identification (ReID) of objects in images is a widely studied topic in computer vision, with significant relevance to various applications. The ReID of players in broadcast videos of team sports is the focus of this study. We specifically focus on identifying the same player in images taken at any given moment during a game from various camera angles. This work varies from other person ReID apps since the same team wears very similar clothes, there are few samples for each identification, and image resolutions are low. One of the hardest parts of object ReID is robust feature representation extraction. Despite the great success of current convolutional neural network-based (CNN) methods, most studies only consider learning representations from images, neglecting long-range dependency. Transformer-based model studies are increasing and yielding encouraging results. Transformers still have trouble extracting features from small objects and visual cues. To address these issues, we enhanced the Swin Transformer with the levering of CNNs. We created a regional feature extraction Swin Transformer (RFES) backbone to increase local feature extraction and small-scale object feature extraction. We also use three loss functions to handle imbalanced data and highlight challenging situations. Re-ranking with k-reciprocal encoding was used in this study's retrieval phase, and its assessment findings were provided. Finally, we conducted experiments on the Market-1501 and SoccerNet-v3 ReID datasets. Experimental results show that the proposed re-ID method reaches rank-1 accuracy of 96.2% with mAP: 89.1 and rank-1 accuracy of 84.1% with mAP: 86.7 on the Market-1501 and SoccerNet-v3 datasets, respectively, outperforming the state-of-the-art approaches.

Elevating healthcare through artificial intelligence: analyzing the abdominal emergencies data set (TR_ABDOMEN_RAD_EMERGENCY) at TEKNOFEST-2022.

OBJECTIVES: The artificial intelligence competition in healthcare at TEKNOFEST-2022 provided a platform to address the complex multi-class classification challenge of abdominal emergencies using computer vision techniques. This manuscript aimed to comprehensively present the methodologies for data preparation, annotation procedures, and rigorous evaluation metrics. Moreover, it was conducted to introduce a meticulously curated abdominal emergencies data set to the researchers. METHODS: The data set underwent a comprehensive central screening procedure employing diverse algorithms extracted from the e-Nabiz (Pulse) and National Teleradiology System of the Republic of Türkiye, Ministry of Health. Full anonymization of the data set was conducted. Subsequently, the data set was annotated by a group of ten experienced radiologists. The evaluation process was executed by calculating F1 scores, which were derived from the intersection over union values between the predicted bounding boxes and the corresponding ground truth (GT) bounding boxes. The establishment of baseline performance metrics involved computing the average of the highest five F1 scores. RESULTS: Observations indicated a progressive decline in F1 scores as the threshold value increased. Furthermore, it could be deduced that class 6 (abdominal aortic aneurysm/dissection) was relatively straightforward to detect compared to other classes, with class 5 (acute diverticulitis) presenting the most formidable challenge. It is noteworthy, however, that if all achieved outcomes for all classes were considered with a threshold of 0.5, the data set's complexity and associated challenges became pronounced. CONCLUSION: This data set's significance lies in its pioneering provision of labels and GT-boxes for six classes, fostering opportunities for researchers. CLINICAL RELEVANCE STATEMENT: The prompt identification and timely intervention in cases of emergent medical conditions hold paramount significance. The handling of patients' care can be augmented, while the potential for errors is minimized, particularly amidst high caseload scenarios, through the application of AI. KEY POINTS: • The data set used in artificial intelligence competition in healthcare (TEKNOFEST-2022) provides a 6-class data set of abdominal CT images consisting of a great variety of abdominal emergencies. • This data set is compiled from the National Teleradiology System data repository of emergency radiology departments of 459 hospitals. • Radiological data on abdominal emergencies is scarce in literature and this annotated competition data set can be a valuable resource for further studies and new AI models.

Artificial Intelligence in Healthcare Competition (TEKNOFEST-2021): Stroke Data Set.

OBJECTIVE: The artificial intelligence competition in healthcare was organized for the first time at the annual aviation, space, and technology festival (TEKNOFEST), Istanbul/Türkiye, in September 2021. In this article, the data set preparation and competition processes were explained in detail; the anonymized and annotated data set is also provided via official website for further research. MATERIALS AND METHODS: Data set recorded over the period covering 2019 and 2020 were centrally screened from the e-Pulse and Teleradiology System of the Republic of Türkiye, Ministry of Health using various codes and filtering criteria. The data set was anonymized. The data set was prepared, pooled, curated, and annotated by 7 radiologists. The training data set was shared with the teams via a dedicated file transfer protocol server, which could be accessed using private usernames and passwords given to the teams under a nondisclosure agreement signed by the representative of each team. RESULTS: The competition consisted of 2 stages. In the first stage, teams were given 192 digital imaging and communications in medicine images that belong to 1 of 3 possible categories namely, hemorrhage, ischemic, or non-stroke. Teams were asked to classify each image as either stroke present or absent. In the second stage of the competition, qualifying 36 teams were given 97 digital imaging and communications in medicine images that contained hemorrhage, ischemia, or both lesions. Among the employed methods, Unet and DeepLabv3 were the most frequently observed ones. CONCLUSION: Artificial intelligence competitions in healthcare offer good opportunities to collect data reflecting various cases and problems. Especially, annotated data set by domain experts is more valuable.

A novel approach for computerized quantitative image analysis of proximal femur bone shape deformities based on the hip joint symmetry.

As a result of most of the bone disorders seen in hip joints, shape deformities occur in the structural form of the hip joint components. Image-based quantitative analysis and assessment of these deformities in bone shapes are very important for the evaluation, treatment, and prognosis of the various hip joint bone disorders. In this article, a novel approach for the image-based computerized quantitative analysis of proximal femur shape deformities is presented. In the proposed approach, shape deformities of the pathological proximal femurs were quantified over the contralateral healthy proximal femur shape structure of the same patient in 2D by taking the hip joint symmetry property of human anatomy into consideration. It is based on the idea that if the right and left proximal femurs in bilateral hip joints are highly symmetrical and also if one of the proximal femurs is healthy and the contralateral one is pathological, the non-overlapping bone shape regions can represent the deformities in pathological proximal femurs when both proximal femurs are registered to overlap each other. In the methodological process of the proposed study, a set of image preprocessing operations was primarily performed on the raw magnetic resonance imaging (MRI) data. Then, the segmented proximal femurs in bilateral hip joint images were automatically aligned with the Iterative Closest Point (ICP) rigid registration method. Following the registration, a set of image postprocessing operations was performed on the images of proximal femurs aligned. In the quantification phase, the bone shape deformities in pathological proximal femurs were quantified simply in terms of the mismatching area in 2D by measuring a shape variation index representing the total bone shape deformity ratio. To evaluate the proposed quantitative shape analysis approach, bilateral hip joints in a total of 13 coronal MRI sections of 13 patients with Legg-Calve-Perthes disease (LCPD) were used. Experimental studies have shown that the proposed approach has quite promising results in the quantitative representation of the pathological proximal femur shape deformities. Furthermore, consistent results have been observed for the Waldenström classification stages of the disease. The shape deformity ratios in pathological proximal femurs were quantified as 9.44% (±1.40), 18.38% (±6.30), 24.73% (±12.42), and 27.66% (±10.41), respectively for the Initial, Fragmentation, Reossification, and Remodelling stages of LCPD with the quantification error rates of 0.29% (±0.16), 0.58% (±0.71), 1.12% (±0.82), and 0.80% (±0.98). Additionally, a mean error rate of 0.65% (±0.68) was observed for the quantified shape deformity ratios of all samples.

Semantic segmentation of the multiform proximal femur and femoral head bones with the deep convolutional neural networks in low quality MRI sections acquired in different MRI protocols.

Medical image segmentation is one of the most crucial issues in medical image processing and analysis. In general, segmentation of the various structures in medical images is performed for the further image analyzes such as quantification, assessment, diagnosis, prognosis and classification. In this paper, a research study for the 2D semantic segmentation of the multiform, both spheric and aspheric, femoral head and proximal femur bones in magnetic resonance imaging (MRI) sections of the patients with Legg-Calve-Perthes disease (LCPD) with the deep convolutional neural networks (CNNs) is presented. In the scope of the proposed study, bilateral hip MRI sections acquired in coronal plane were used. The main characteristic of the MRI sections that were used is to be low quality images which were obtained in different MRI protocols by using 3 different MRI scanners with 1.5 T imaging capability. In performance evaluations, promising segmentation results were achieved with deep CNNs in low quality MRI sections acquired in different MRI protocols. A success rate about 90% was observed in semantic segmentation of the multiform femoral head and proximal femur bones in a total of 194 MRI sections obtained from 33 MRI sequences of 13 patients with deep CNNs.