OBJECTIVE: The incidence of facial fractures is on the rise globally, yet limited studies are addressing the diverse forms of facial fractures present in 3D images. In particular, due to the nature of the facial fracture, the direction in which the bone fractures vary, and there is no clear outline, it is difficult to determine the exact location of the fracture in 2D images. Thus, 3D image analysis is required to find the exact fracture area, but it needs heavy computational complexity and expensive pixel-wise labeling for supervised learning. In this study, we tackle the problem of reducing the computational burden and increasing the accuracy of fracture localization by using a weakly-supervised object localization without pixel-wise labeling in a 3D image space.
METHODS: We propose a Very Fast, High-Resolution Aggregation 3D Detection CAM (VFHA-CAM) model, which can detect various facial fractures. To better detect tiny fractures, our model uses high-resolution feature maps and employs Ablation CAM to find an exact fracture location without pixel-wise labeling, where we use a rough fracture image detected with 3D box-wise labeling. To this end, we extract important features and use only essential features to reduce the computational complexity in 3D image space.
RESULTS: Experimental findings demonstrate that VFHA-CAM surpasses state-of-the-art 2D detection methods by up to 20% in sensitivity/person and specificity/person, achieving sensitivity/person and specificity/person scores of 87% and 85%, respectively. In addition, Our VFHA-CAM reduces location analysis time to 76 s without performance degradation compared to a simple Ablation CAM method that takes more than 20 min.
CONCLUSIONS: This study introduces a novel weakly-supervised object localization approach for bone fracture detection in 3D facial images. The proposed method employs a 3D detection model, which helps detect various forms of facial bone fractures accurately. The CAM algorithm adopted for fracture area segmentation within a 3D fracture detection box is key in quickly informing medical staff of the exact location of a facial bone fracture in a weakly-supervised object localization. In addition, we provide 3D visualization so that even non-experts unfamiliar with 3D CT images can identify the fracture status and location.

Deep learning has been widely used in ultrasound image analysis, and it also benefits kidney ultrasound interpretation and diagnosis. However, the importance of ultrasound image resolution often goes overlooked within deep learning methodologies. In this study, we integrate the ultrasound image resolution into a convolutional neural network and explore the effect of the resolution on diagnosis of kidney tumors. In the process of integrating the image resolution information, we propose two different approaches to narrow the semantic gap between the features extracted by the neural network and the resolution features. In the first approach, the resolution is directly concatenated with the features extracted by the neural network. In the second approach, the features extracted by the neural network are first dimensionally reduced and then combined with the resolution features to form new composite features. We compare these two approaches incorporating the resolution with the method without incorporating the resolution on a kidney tumor dataset of 926 images consisting of 211 images of benign kidney tumors and 715 images of malignant kidney tumors. The area under the receiver operating characteristic curve (AUC) of the method without incorporating the resolution is 0.8665, and the AUCs of the two approaches incorporating the resolution are 0.8926 (P < 0.0001) and 0.9135 (P < 0.0001) respectively. This study has established end-to-end kidney tumor classification systems and has demonstrated the benefits of integrating image resolution, showing that incorporating image resolution into neural networks can more accurately distinguish between malignant and benign kidney tumors in ultrasound images.

Objective:To build a VGG-based computer-aided diagnostic model for chronic sinusitis and evaluate its efficacy. Methods:①A total of 5 000 frames of diagnosed sinus CT images were collected. The normal group consisted of 1 000 frames（250 frames each of maxillary sinus, frontal sinus, septal sinus, and pterygoid sinus）, while the abnormal group consisted of 4 000 frames（1 000 frames each of maxillary sinusitis, frontal sinusitis, septal sinusitis, and pterygoid sinusitis）. ②The models were trained and simulated to obtain five classification models for the normal group, the pteroid sinusitis group, the frontal sinusitis group, the septal sinusitis group and the maxillary sinusitis group, respectively. The classification efficacy of the models was evaluated objectively in six dimensions: accuracy, precision, sensitivity, specificity, interpretation time and area under the ROC curve（AUC）. ③Two hundred randomly selected images were read by the model with three groups of physicians（low, middle and high seniority） to constitute a comparative experiment. The efficacy of the model was objectively evaluated using the aforementioned evaluation indexes in conjunction with clinical analysis. Results:①Simulation experiment: The overall recognition accuracy of the model is 83.94%, with a precision of 89.52%, sensitivity of 83.94%, specificity of 95.99%, and the average interpretation time of each frame is 0.2 s. The AUC for sphenoid sinusitis was 0.865（95%CI 0.849-0.881）, for frontal sinusitis was 0.924（0.991-0.936）, for ethmoidoid sinusitis was 0.895（0.880-0.909）, and for maxillary sinusitis was 0.974（0.967-0.982）. ②Comparison experiment: In terms of recognition accuracy, the model was 84.52%, while the low-seniority physicians group was 78.50%, the middle-seniority physicians group was 80.50%, and the seniority physicians group was 83.50%; In terms of recognition accuracy, the model was 85.67%, the low seniority physicians group was 79.72%, the middle seniority physicians group was 82.67%, and the high seniority physicians group was 83.66%. In terms of recognition sensitivity, the model was 84.52%, the low seniority group was 78.50%, the middle seniority group was 80.50%, and the high seniority group was 83.50%. In terms of recognition specificity, the model was 96.58%, the low-seniority physicians group was 94.63%, the middle-seniority physicians group was 95.13%, and the seniority physicians group was 95.88%. In terms of time consumption, the average image per frame of the model is 0.20 s, the average image per frame of the low-seniority physicians group is 2.35 s, the average image per frame of the middle-seniority physicians group is 1.98 s, and the average image per frame of the senior physicians group is 2.19 s. Conclusion:This study demonstrates the potential of a deep learning-based artificial intelligence diagnostic model for chronic sinusitis to classify and diagnose chronic sinusitis; the deep learning-based artificial intelligence diagnosis model for chronic sinusitis has good classification performance and high diagnostic efficacy.
目的：搭建基于VGG的慢性鼻窦炎计算机辅助诊断模型，并评价其效能。 方法：①收集5 000帧已确诊的鼻窦CT图像，将其分为正常组1 000帧图像（其中，正常的上颌窦、额窦、筛窦、蝶窦影像图像各250帧）及异常组4 000帧图像（其中，上颌窦炎、额窦炎、筛窦炎、蝶窦炎影像图像各1 000帧），对图像进行大小归一化及分割预处理；②训练模型并对其进行仿真实验，分别得到正常组，蝶窦炎组，额窦炎组，筛窦炎组以及上颌窦炎组5个分类模型，从准确度、精确度、灵敏度、特异度、判读时间及ROC曲线下面积（AUC）6个维度，客观评价模型的分类效能；③随机选取200帧图像，通过模型与低年资医师组、中年资医师组、高年资医师组分别阅片构成对比试验，结合临床通过以上评价指标客观评价模型的效能。 结果：①仿真实验：整个模型的识别准确度为83.94%，精确度为89.52%，灵敏度为83.94%，特异度为95.99%，平均每帧图像判读时间为0.20 s；蝶窦炎的AUC为0.865（95%CI 0.849～0.881），额窦炎的AUC为0.924（0.911～0.936），筛窦炎的AUC为0.895（0.880～0.909），上颌窦炎的AUC为0.974（0.967～0.982）。②对比实验：在识别准确度上，模型为84.52%，低年资医师组为78.5%、中年资医师组为80.5%，高年资医师组为83.5%；在识别精确度上，模型为85.67%，低年资医师组为79.72%，中年资医师组为82.67%，高年资医师组为83.66%；在识别灵敏度上，模型为84.52%，低年资医师组为78.50%，中年资医师组为80.50%，高年资医师组为83.50%；在识别特异度上，模型为96.58%，低年资医师组为94.63%，中年资医师组为95.13%，高年资医师组为95.88%；在耗时上，模型平均每帧图像为0.20 s，低年资医师组平均每帧图像为2.35 s，中年资医师组平均每帧图像为1.98 s，高年资医师组平均每帧图像为2.19 s。 结论：本研究强调了基于深度学习的慢性鼻窦炎人工智能诊断模型分类诊断慢性鼻窦炎的可能性；基于深度学习的慢性鼻窦炎人工智能诊断模型分类性能好，具有较高的诊断效能。.

Objective.The trend in the medical field is towards intelligent detection-based medical diagnostic systems. However, these methods are often seen as \'black boxes\' due to their lack of interpretability. This situation presents challenges in identifying reasons for misdiagnoses and improving accuracy, which leads to potential risks of misdiagnosis and delayed treatment. Therefore, how to enhance the interpretability of diagnostic models is crucial for improving patient outcomes and reducing treatment delays. So far, only limited researches exist on deep learning-based prediction of spontaneous pneumothorax, a pulmonary disease that affects lung ventilation and venous return.Approach.This study develops an integrated medical image analysis system using explainable deep learning model for image recognition and visualization to achieve an interpretable automatic diagnosis process.Main results.The system achieves an impressive 95.56% accuracy in pneumothorax classification, which emphasizes the significance of the blood vessel penetration defect in clinical judgment.Significance.This would lead to improve model trustworthiness, reduce uncertainty, and accurate diagnosis of various lung diseases, which results in better medical outcomes for patients and better utilization of medical resources. Future research can focus on implementing new deep learning models to detect and diagnose other lung diseases that can enhance the generalizability of this system.

Recent advancements in computational intelligence, deep learning, and computer-aided detection have had a significant impact on the field of medical imaging. The task of image segmentation, which involves accurately interpreting and identifying the content of an image, has garnered much attention. The main objective of this task is to separate objects from the background, thereby simplifying and enhancing the significance of the image. However, existing methods for image segmentation have their limitations when applied to certain types of images. This survey paper aims to highlight the importance of image segmentation techniques by providing a thorough examination of their advantages and disadvantages. The accurate detection of cancer regions in medical images is crucial for ensuring effective treatment. In this study, we have also extensive analysis of Computer-Aided Diagnosis (CAD) systems for cancer identification, with a focus on recent research advancements. The paper critically assesses various techniques for cancer detection and compares their effectiveness. Convolutional neural networks (CNNs) have attracted particular interest due to their ability to segment and classify medical images in large datasets, thanks to their capacity for self- learning and decision-making.

Chest X-ray (CXR) imaging is widely employed by radiologists to diagnose thoracic diseases. Recently, many deep learning techniques have been proposed as computer-aided diagnostic (CAD) tools to assist radiologists in minimizing the risk of incorrect diagnosis. From an application perspective, these models have exhibited two major challenges: (1) They require large volumes of annotated data at the training stage and (2) They lack explainable factors to justify their outcomes at the prediction stage. In the present study, we developed a class activation mapping (CAM)-based ensemble model, called Ensemble-CAM, to address both of these challenges via weakly supervised learning by employing explainable AI (XAI) functions. Ensemble-CAM utilizes class labels to predict the location of disease in association with interpretable features. The proposed work leverages ensemble and transfer learning with class activation functions to achieve three objectives: (1) minimizing the dependency on strongly annotated data when locating thoracic diseases, (2) enhancing confidence in predicted outcomes by visualizing their interpretable features, and (3) optimizing cumulative performance via fusion functions. Ensemble-CAM was trained on three CXR image datasets and evaluated through qualitative and quantitative measures via heatmaps and Jaccard indices. The results reflect the enhanced performance and reliability in comparison to existing standalone and ensembled models.

OBJECTIVE: Current automatic electrocardiogram (ECG) diagnostic systems could provide classification outcomes but often lack explanations for these results. This limitation hampers their application in clinical diagnoses. Previous supervised learning could not highlight abnormal segmentation output accurately enough for clinical application without manual labeling of large ECG datasets.
METHODS: In this study, we present a multi-instance learning framework called MA-MIL, which has designed a multi-layer and multi-instance structure that is aggregated step by step at different scales. We evaluated our method using the public MIT-BIH dataset and our private dataset.
RESULTS: The results show that our model performed well in both ECG classification output and heartbeat level, sub-heartbeat level abnormal segment detection, with accuracy and F1 scores of 0.987 and 0.986 for ECG classification and 0.968 and 0.949 for heartbeat level abnormal detection, respectively. Compared to visualization methods, the IoU values of MA-MIL improved by at least 17 % and at most 31 % across all categories.
CONCLUSIONS: MA-MIL could accurately locate the abnormal ECG segment, offering more trustworthy results for clinical application.

Brain magnetic resonance imaging (MRI) scans are available in a wide variety of sequences, view planes, and magnet strengths. A necessary preprocessing step for any automated diagnosis is to identify the MRI sequence, view plane, and magnet strength of the acquired image. Automatic identification of the MRI sequence can be useful in labeling massive online datasets used by data scientists in the design and development of computer aided diagnosis (CAD) tools. This paper presents a deep learning (DL) approach for brain MRI sequence and view plane identification using scans of different data types as input. A 12-class classification system is presented for commonly used MRI scans, including T1, T2-weighted, proton density (PD), fluid attenuated inversion recovery (FLAIR) sequences in axial, coronal and sagittal view planes. Multiple online publicly available datasets have been used to train the system, with multiple infrastructures. MobileNet-v2 offers an adequate performance accuracy of 99.76% with unprocessed MRI scans and a comparable accuracy with skull-stripped scans and has been deployed in a tool for public use. The tool has been tested on unseen data from online and hospital sources with a satisfactory performance accuracy of 99.84 and 86.49%, respectively.

OBJECTIVE: Malocclusion has emerged as a burgeoning global public health concern. Individuals with an anterior crossbite face an elevated risk of exhibiting characteristics such as a concave facial profile, negative overjet, and poor masticatory efficiency. In response to this issue, we proposed a convolutional neural network (CNN)-based model designed for the automated detection and classification of intraoral images and videos.
METHODS: A total of 1865 intraoral images were included in this study, 1493 (80%) of which were allocated for training and 372 (20%) for testing the CNN. Additionally, we tested the models on 10 videos, spanning a cumulative duration of 124 seconds. To assess the performance of our predictions, metrics including accuracy, sensitivity, specificity, precision, F1-score, area under the precision-recall (AUPR) curve, and area under the receiver operating characteristic (ROC) curve (AUC) were employed.
RESULTS: The trained model exhibited commendable classification performance, achieving an accuracy of 0.965 and an AUC of 0.986. Moreover, it demonstrated superior specificity (0.992 vs. 0.978 and 0.956, P < 0.05) in comparison to assessments by two orthodontists. Conversely, the CNN model displayed diminished sensitivity (0.89 vs. 0.96 and 0.92, P < 0.05) relative to the orthodontists. Notably, the CNN model accomplished a perfect classification rate, successfully identifying 100% of the videos in the test set.
CONCLUSIONS: The deep learning (DL) model exhibited remarkable classification accuracy in identifying anterior crossbite through both intraoral images and videos. This proficiency holds the potential to expedite the detection of severe malocclusions, facilitating timely classification for appropriate treatment and, consequently, mitigating the risk of complications.

OBJECTIVE: The effective segmentation of esophageal squamous carcinoma lesions in CT scans is significant for auxiliary diagnosis and treatment. However, accurate lesion segmentation is still a challenging task due to the irregular form of the esophagus and small size, the inconsistency of spatio-temporal structure, and low contrast of esophagus and its peripheral tissues in medical images. The objective of this study is to improve the segmentation effect of esophageal squamous cell carcinoma lesions.
METHODS: It is critical for a segmentation network to effectively extract 3D discriminative features to distinguish esophageal cancers from some visually closed adjacent esophageal tissues and organs. In this work, an efficient HRU-Net architecture (High-Resolution U-Net) was exploited for esophageal cancer and esophageal carcinoma segmentation in CT slices. Based on the idea of localization first and segmentation later, the HRU-Net locates the esophageal region before segmentation. In addition, an Resolution Fusion Module (RFM) was designed to integrate the information of adjacent resolution feature maps to obtain strong semantic information, as well as preserve the high-resolution features.
RESULTS: Compared with the other five typical methods, the devised HRU-Net is capable of generating superior segmentation results.
CONCLUSIONS: Our proposed HRU-NET improves the accuracy of segmentation for squamous esophageal cancer. Compared to other models, our model performs the best. The designed method may improve the efficiency of clinical diagnosis of esophageal squamous cell carcinoma lesions.