Characteristics Of Medical Images Research Articles

Core-Periphery Multi-Modality Feature Alignment for Zero-Shot Medical Image Analysis.

Multi-modality learning, exemplified by the language-image pair pre-trained CLIP model, has demonstrated remarkable performance in enhancing zero-shot capabilities and has gained significant attention recently. However, simply applying language-image pre-trained CLIP to medical image analysis encounters substantial domain shifts, resulting in severe performance degradation due to inherent disparities between natural (non-medical) and medical image characteristics. To address this challenge and uphold or even enhance CLIP's zero-shot capability in medical image analysis, we develop a novel approach, Core-Periphery feature alignment for CLIP (CP-CLIP), to model medical images and corresponding clinical text jointly. To achieve this, we design an auxiliary neural network whose structure is organized by the core-periphery (CP) principle. This auxiliary CP network not only aligns medical image and text features into a unified latent space more efficiently but also ensures alignment driven by principles of brain network organization. In this way, our approach effectively mitigates and further enhances CLIP's zero-shot performance in medical image analysis. More importantly, the proposed CP-CLIP exhibits excellent explanatory capability, enabling the automatic identification of critical disease-related regions in clinical analysis. Extensive experiments and evaluation across five public datasets covering different diseases underscore the superiority of our CP-CLIP in zero-shot medical image prediction and critical features detection, showing its promising utility in multimodal feature alignment in current medical applications.

The Current Progress of Artificial Intelligence in Approach to Thyroid Nodules: A Narrative Review

Background: Artificial intelligence (AI) can play a significant role in the future of thyroidology. Thyroid nodules are common conditions that may benefit from AI through more accurate and efficient diagnosis, risk stratification, and medical or surgical management. Objective: This paper aims to review the latest developments in AI applications for diagnosing and managing thyroid nodules and cancers. Methods: English full-text articles published in the PubMed and Google Scholar databases from January 2014 to March 2024 were collected and reviewed to provide a comprehensive understanding of the topic. A total of 45 studies were selected based on relevance, robust methodology, statistical significance, and broader topic coverage. Results: Artificial intelligence has emerged as a powerful tool for managing thyroid nodules. First, several studies have demonstrated how AI-powered ultrasound interpretation enhances the diagnosis and classification of nodules while reducing the need for invasive fine-needle aspiration (FNA) biopsies. Second, AI significantly improves the cytopathological differentiation between benign and malignant thyroid nodules by minimizing reliance on pathologists' expertise and implementing standardized diagnostic criteria. When cytopathology is inconclusive, AI also aids in identifying molecular markers from omics data, distinguishing between normal and cancerous cells. Moreover, AI tools have been developed for prognosis assessment, predicting distant metastasis, recurrence, and surveillance by integrating medical imaging features with molecular and clinical factors. Additionally, some AI tools are designed for intraoperative evaluation, improving surgical techniques and reducing complications during thyroidectomy. In non-surgical treatments, several models have been developed to optimize therapeutic doses of radioactive iodine (RAI) and predict the outcomes of new drug formulations. Conclusions: Artificial intelligence has the potential to assist physicians in accurate thyroid nodule diagnosis, classification, decision-making, optimizing treatment strategies, and improving patient outcomes. However, there are still limitations to this technology. Artificial intelligence-driven tools require further advancements before they can be fully integrated into clinical practice and replace specialists.

An intelligent bone age assessment model incorporating multilayer superimposed texture enhancement and the China-05 attention mechanism

In the process of intelligent bone age assessment for Chinese ethnicity, generalized convolutional network models are not well targeted in extracting specific features in medical skeletal images and lack specificity in training and predicting skeletal developmental features in different ethnicities. This study aims to propose a hybrid improved deep residual network model, ZH05-DL-ResNet50, focusing on intelligent bone age assessment for Chinese ethnicity. In the process of building the ZH05-DL-ResNet50 model. First, multiple overlapping texture enhancement layers are introduced through combinatorial optimization, which can better characterize the global features of hand bone radiographs while using less texture information, reducing the interference of redundant information and freeing up computational power; second, the China-05′s spatial focusing mechanism is designed so that the model can intelligently focus on the region of interest, efficiently locate and automatically learn key image information; Finally, the model can be used for intelligent bone age assessment of Chinese ethnicity by constructing a 50-layer residual depth network, the superposition enhancement layer and the focusing mechanism are fused to form the ZH05-DL-ResNet50 model, which mitigates the problem of gradient disappearance and explosion caused by the increase of network depth, and reduces the information loss and depletion of the multi-layer texture superposition features in the focus region of interest. The training was performed using the Chinese ethnographic dataset provided and created by Tsinghua Changgeng Hospital, where 10 % of the radiographs were used as the test set and the rest as the training and validation sets. The overall performance of the models was estimated by comparing the average accuracy and loss values of different models, and the overall performance between model training and model design was evaluated using the data distribution (Distribution) and histogram of bias weight distribution (Histogram). The results showed that the average accuracy of the bone age estimation model was 98.1 % and the average error on the test set was 0.312 years. Evaluation of the model visualization and accuracy comparison showed that the model performed well and was more appropriate and accurate than other bone age estimation models. Therefore, the first intelligent bone age assessment model for Chinese ethnicity, ZH05-DL-ResNet50, was created, which differs from the untargeted nature of other intelligent bone age assessment models, and is used to train and predict hand bone radiographs more appropriately and accurately for Chinese ethnicity.

Correlating Histopathological Microscopic Images of Creutzfeldt–Jakob Disease with Clinical Typology Using Graph Theory and Artificial Intelligence

Creutzfeldt–Jakob disease (CJD) is a rare, degenerative, and fatal brain disorder caused by abnormal proteins called prions. This research introduces a novel approach combining AI and graph theory to analyze histopathological microscopic images of brain tissues affected by CJD. The detection and quantification of spongiosis, characterized by the presence of vacuoles in the brain tissue, plays a crucial role in aiding the accurate diagnosis of CJD. The proposed methodology employs image processing techniques to identify these pathological features in high-resolution medical images. By developing an automatic pipeline for the detection of spongiosis, we aim to overcome some limitations of manual feature extraction. The results demonstrate that our method correctly identifies and characterize spongiosis and allows the extraction of features that will help to better understand the spongiosis patterns in different CJD patients.

The value of 18F-fluorodeoxyglucose positron emission tomography-based radiomics in non-small cell lung cancer

ABSTRACT Currently, the second most commonly diagnosed cancer in the world is lung cancer, and 85% of cases are non-small cell lung cancer (NSCLC). With growing knowledge of oncogene drivers and cancer immunology, several novel therapeutics have emerged to improve the prognostic outcomes of NSCLC. However, treatment outcomes remain diverse, and an accurate tool to achieve precision medicine is an unmet need. Radiomics, a method of extracting medical imaging features, is promising for precision medicine. Among all radiomic tools, 18F-fluorodeoxyglucose positron emission tomography (18F-FDG PET)-based radiomics provides distinct information on glycolytic activity and heterogeneity. In this review, we collected relevant literature from PubMed and summarized the various applications of 18F-FDG PET-derived radiomics in improving the detection of metastasis, subtyping histopathologies, characterizing driver mutations, assessing treatment response, and evaluating survival outcomes of NSCLC. Furthermore, we reviewed the values of 18F-FDG PET-based deep learning. Finally, several challenges and caveats exist in the implementation of 18F-FDG PET-based radiomics for NSCLC. Implementing 18F-FDG PET-based radiomics in clinical practice is necessary to ensure reproducibility. Moreover, basic studies elucidating the underlying biological significance of 18F-FDG PET-based radiomics are lacking. Current inadequacies hamper immediate clinical adoption; however, radiomic studies are progressively addressing these issues. 18F-FDG PET-based radiomics remains an invaluable and indispensable aspect of precision medicine for NSCLC.

RESEARCH ON RETINEX ENHANCEMENT METHOD FOR VASCULAR MEDICAL IMAGES

To improve the feature extraction performance in vascular medical images, a multi-scale Retinex medical image enhancement algorithm based on a four-step method is proposed. This method is based on the multi-scale Retinex enhancement method, which divides the entire enhancement process into four steps: The first step is to reduce the overall complexity of the enhancement process through block processing; second, perform Retinex enhancement in each sub-block image; the third step is to optimize the connection area of each sub-block; and the fourth step is to highlight detailed features through gradient filtering. During the experiment, three different images were enhanced. The experimental results show that the proposed method can be used in enhancing the vascular features in medical images and be extended to other application fields. This method can effectively suppress the blurring and halo of the original image, while achieving enhancement effects and better protecting detailed features. From the four sets of indicators, the proposed method has shown more significant improvement than both reference methods. Regarding information entropy, the doubles have increased by a maximum of 96.2% and a minimum of 30.3%.

Reliable segmentation of multiple lesions from medical images.

Focusing on the complicated pathological features, such as blurred boundaries, severe scale differences between symptoms, and background noise interference, we aim to enhance the reliability of multiple lesions joint segmentation from medicalimages. Propose a novel reliable multi-scale wavelet-enhanced transformer network, which can provide accurate segmentation results with reliabilityassessment. Focusing on enhancing the model's capability to capture intricate pathological features in medical images, this work introduces a novel segmentation backbone. The backbone integrates a wavelet-enhanced feature extractor network and incorporates a multi-scale transformer module developed within the scope of this work. Simultaneously, to enhance the reliability of segmentation outcomes, a novel uncertainty segmentation head is proposed. This segmentation head is rooted in the SL, contributing to the generation of final segmentation results along with an associated overall uncertainty evaluation scoremap. Comprehensive experiments are conducted on the public database of AI-Challenge 2018 for retinal edema lesions segmentation and the segmentation of Thoracic Organs at Risk in CT images. The experimental results highlight the superior segmentation accuracy and heightened reliability achieved by the proposed method in comparison to other state-of-the-art segmentationapproaches. Unlike previous segmentation methods, the proposed approach can produce reliable segmentation results with an estimated uncertainty and higher accuracy, enhancing the overall reliability of the model. The code will be release on https://github.com/LooKing9218/ReMultiSeg.

Expansive Receptive Field and Local Feature Extraction Network: Advancing Multiscale Feature Fusion for Breast Fibroadenoma Segmentation in Sonography.

Fibroadenoma is a common benign breast disease that affects women of all ages. Early diagnosis can greatly improve the treatment outcomes and reduce the associated pain. Computer-aided diagnosis (CAD) has great potential to improve diagnosis accuracy and efficiency. However, its application in sonography is limited. A network that utilizes expansive receptive fields and local information learning was proposed for the accurate segmentation of breast fibroadenomas in sonography. The architecture comprises the Hierarchical Attentive Fusion module, which conducts local information learning through channel-wise and pixel-wise perspectives, and the Residual Large-Kernel module, which utilizes multiscale large kernel convolution for global information learning. Additionally, multiscale feature fusion in both modules was included to enhance the stability of our network. Finally, an energy function and a data augmentation method were incorporated to fine-tune low-level features of medical images and improve data enhancement. The performance of our model is evaluated using both our local clinical dataset and a public dataset. Mean pixel accuracy (MPA) of 93.93% and 86.06% and mean intersection over union (MIOU) of 88.16% and 73.19% were achieved on the clinical and public datasets, respectively. They are significantly improved over state-of-the-art methods such as SegFormer (89.75% and 78.45% in MPA and 83.26% and 71.85% in MIOU, respectively). The proposed feature extraction strategy, combining local pixel-wise learning with an expansive receptive field for global information perception, demonstrates excellent feature learning capabilities. Due to this powerful and unique local-global feature extraction capability, our deep network achieves superior segmentation of breast fibroadenoma in sonography, which may be valuable in early diagnosis.

Current landscape of primary small bowel leiomyosarcoma: cases report and a decade of insights

The incidence of leiomyosarcoma (LMS) is about 4–5/100,000 individuals per year. LMSs occurring in the small bowel are even rarer, and their preoperative diagnosis is very difficult. We described two patients with pathologically confirmed small bowel LMS and analyzed their clinical and medical imaging features. Similar cases reported in English in Pubmed database over the past decade were reviewed and summarized. These tumors were categorized by the growth direction and relationship with the intestinal lumen into three types: intraluminal (n = 10), intermural (n = 3), and extraluminal (n = 7). Notably, among the three types of LMS, the intramural leiomyosarcoma stands out as a noteworthy subtype. Emerging evidence suggests that smaller tumor size (< 5 cm) and the intraluminal type may serve as favorable prognostic indicators, while the extraluminal type is associated with relatively poor prognosis. Furthermore, the integration of imaging features with CA125 and LDH biomarkers holds promise for potential diagnostic value in LMS.

Optimal strategies for modeling anatomy in a hybrid intelligence framework for auto-segmentation of organs.

Organ segmentation is a fundamental requirement in medical image analysis. Many methods have been proposed over the past 6 decades for segmentation. A unique feature of medical images is the anatomical information hidden within the image itself. To bring natural intelligence (NI) in the form of anatomical information accumulated over centuries into deep learning (DL) AI methods effectively, we have recently introduced the idea of hybrid intelligence (HI) that combines NI and AI and a system based on HI to perform medical image segmentation. This HI system has shown remarkable robustness to image artifacts, pathology, deformations, etc. in segmenting organs in the Thorax body region in a multicenter clinical study. The HI system utilizes an anatomy modeling strategy to encode NI and to identify a rough container region in the shape of each object via a non-DL-based approach so that DL training and execution are applied only to the fuzzy container region. In this paper, we introduce several advances related to modeling of the NI component so that it becomes substantially more efficient computationally, and at the same time, is well integrated with the DL portion (AI component) of the system. We demonstrate a 9-40 fold computational improvement in the auto-segmentation task for radiation therapy (RT) planning via clinical studies obtained from 4 different RT centers, while retaining state-of-the-art accuracy of the previous system in segmenting 11 objects in the Thorax body region.

Enhancing Diagnostic Images to Improve the Performance of the Segment Anything Model in Medical Image Segmentation.

Medical imaging serves as a crucial tool in current cancer diagnosis. However, the quality of medical images is often compromised to minimize the potential risks associated with patient image acquisition. Computer-aided diagnosis systems have made significant advancements in recent years. These systems utilize computer algorithms to identify abnormal features in medical images, assisting radiologists in improving diagnostic accuracy and achieving consistency in image and disease interpretation. Importantly, the quality of medical images, as the target data, determines the achievable level of performance by artificial intelligence algorithms. However, the pixel value range of medical images differs from that of the digital images typically processed via artificial intelligence algorithms, and blindly incorporating such data for training can result in suboptimal algorithm performance. In this study, we propose a medical image-enhancement scheme that integrates generic digital image processing and medical image processing modules. This scheme aims to enhance medical image data by endowing them with high-contrast and smooth characteristics. We conducted experimental testing to demonstrate the effectiveness of this scheme in improving the performance of a medical image segmentation algorithm.

Fully automated diagnosis of thyroid nodule ultrasound using brain-inspired inference

The interpretability of artificial intelligence (AI) based medical diagnostic systems is crucial to make the diagnosis adequately convincible. Deep learning has been extensively investigated and utilized in the area of medical assistance diagnosis in recent decades due to its outstanding performance and objective prediction. However, a huge semantic chasm dividing clinicians and unexplainable deep models emerges. Here we design a brain-inspired inference framework from medical images to explainable features, then to the final diagnostic conclusions. The fast thinking module is responsible for recognizing medical features in ultrasound (US) images, and the slow-thinking module builds a model for inferring from medical features to diagnostic results by constructing a knowledge graph of medical features with tensor decomposition. The whole model infers through intuition and thinking like a human being, and gives the recognized medical image features while inferring the diagnosis, which greatly improves the interpretability of the model. We conducted studies on thyroid cancer diagnoses using US images. The American College of Radiology (ACR) Thyroid Imaging Reporting and Data System (TI-RADS) characteristics are employed as medical features describing thyroid nodules. Our brain-inspired medical inference framework outperforms commonly used deep learning algorithms, with an AUC score of 0.963 (95% confidential interval (CI)=0.923–1.000) for thyroid US image diagnosis. Results indicate that our framework improves diagnostic objectivity and interpretability while providing performance that is better than deep models. Our proposed brain-inspired medical inference framework could improve the efficiency of diagnosis and our technique is performant, objective and interpretable.

A multi-granularity ensemble algorithm for medical image classification based on broad learning system

Medical image classification is an essential task in the fields of computer-aided diagnosis and medical image analysis. In recent years, researchers have made extensive work on medical image classification by computer vision techniques. However, most of the current work is based on deep learning methods, which still suffer from expensive hardware resources, long time consuming and a lot of parameters to be optimized. In this paper, a multi-granularity ensemble algorithm for medical image classification based on broad learning system is proposed, which is an end-to-end lightweight model. On the one hand, the proposed method is designed to address the problem of weak image feature learning ability of broad learning system. The convolution module with fixed weights based on transfer learning is introduced as a feature extractor to extract fusion features of medical images. On the other hand, the multi-granularity ensemble framework is proposed, which learn the fusion features of medical images from fine-grained to coarse-grained respectively, and the prediction results at different granularity levels are integrated by ensemble learning. In this way, the bottom local features can be sufficiently considered, while the global features can also be taken into account. The experimental results show that on the MedMNIST dataset (containing 10 sub-datasets), the proposed method can shorten the training time by tens of times while having similar accuracy to deep convolutional neural networks. On the ChestXRay2017 dataset, the proposed method can achieve an accuracy of 92.5%, and the training time is also significantly better than other methods.

RED-Net: Residual and Enhanced Discriminative Network for Image Steganalysis in the Internet of Medical Things and Telemedicine.

Internet of Medical Things (IoMT) and telemedicine technologies utilize computers, communications, and medical devices to facilitate off-site exchanges between specialists and patients, specialists, and medical staff. If the information communicated in IoMT is illegally steganography, tampered or leaked during transmission and storage, it will directly impact patient privacy or the consultation results with possible serious medical incidents. Steganalysis is of great significance for the identification of medical images transmitted illegally in IoMT and telemedicine. In this article, we propose a Residual and Enhanced Discriminative Network (RED-Net) for image steganalysis in the internet of medical things and telemedicine. RED-Net consists of a steganographic information enhancement module, a deep residual network, and steganographic information discriminative mechanism. Specifically, a steganographic information enhancement module is adopted by the RED-Net to boost the illegal steganographic signal in texturally complex high-dimensional medical image features. A deep residual network is utilized for steganographic feature extraction and compression. A steganographic information discriminative mechanism is employed by the deep residual network to enable it to recalibrate the steganographic features and drop high-frequency features that are mistaken for steganographic information. Experiments conducted on public and private datasets with data hiding payloads ranging from 0.1bpp/bpnzac-0.5bpp/bpnzac in the spatial and JPEG domain led to RED-Net's steganalysis error PE in the range of 0.0732-0.0010 and 0.231-0.026, respectively. In general, qualitative and quantitative results on public and private datasets demonstrate that the RED-Net outperforms 8 state-of-art steganography detectors.

APPLICATION OF NEW RETRIEVAL METHODS IN MICROSCOPIC MEDICAL IMAGES AND CT IMAGES

Medical image retrieval is of great significance for forming correct medical judgments during the diagnosis and treatment process. In response to the fact that there are rich types of features in medical images, this paper constructs a multi-feature fusion model and applies it to medical image retrieval. This multi-feature fusion model utilizes three features. Color information is characterized using three different moments, texture information is obtained using the LBP (Local Binary Pattern) operator, and shape information is described using Hu moments. After collecting three features of medical images, they are fused into a similarity measure model through a self-set weight structure for comparison in medical image retrieval. Experiments show that our method has satisfactory image retrieval result for medical images such as endoscopic images and computed tomography (CT) images. Despite the continuous improvement of recall indicators, the proposed method still has high precision.

Preoperative multimodal ultrasonic imaging in a case of Peutz-Jeghers syndrome complicated by atypical lobular endocervical glandular hyperplasia: a case report and literature review

BackgroundPeutz-Jeghers syndrome (PJS), an autosomal dominant multiple cancerous disorder, is clinically characterized by mucocutaneous macules and multiple gastrointestinal hamartomatous polyps. Gastric-type endocervical adenocarcinoma (G-EAC), a special subtype of cervical adenocarcinoma with non-specific symptoms and signs, is known to occur in approximately 11% of female patients with PJS.Case presentationHere, we report a case of PJS in a 24-year-old female with multiple mucocutaneous black macules who complained of vaginal discharge and menorrhagia. Moreover, we first described the multimodal ultrasonographical manifestations of PJS-correlated G-EAC. The three-dimensional reconstructed view of G-EAC on 3D realisticVue exhibited a distinctive “cosmos pattern” resembling features on magnetic resonance imaging, and the contrast-enhanced ultrasound displayed a “quick-up and slow-down” pattern of the solid components inside the mixed cervical echoes. We reported the multimodal ultrasonographical characteristics of a case of PJS-related G-EAC, as well as reviewed PJS-related literature and medical imaging features and clinical characteristics of G-EAC to provide insight into the feasibility and potential of utilizing multimodal ultrasonography for the diagnosis of G-EAC.ConclusionsMultimodal ultrasound can visualize morphological features, solid components inside, and blood supplies of the G-EAC lesion and distinguish the G-EAC lesion from normal adjacent tissues. This facilitates preoperative diagnosis and staging of PJS-related G-EAC, thereby aiding subsequent health and reproductive management for patients with PJS.

Research on automatic generation of multimodal medical image reports based on memory driven

The task of automatic generation of medical image reports faces various challenges, such as diverse types of diseases and a lack of professionalism and fluency in report descriptions. To address these issues, this paper proposes a multimodal medical imaging report based on memory drive method (mMIRmd). Firstly, a hierarchical vision transformer using shifted windows (Swin-Transformer) is utilized to extract multi-perspective visual features of patient medical images, and semantic features of textual medical history information are extracted using bidirectional encoder representations from transformers (BERT). Subsequently, the visual and semantic features are integrated to enhance the model's ability to recognize different disease types. Furthermore, a medical text pre-trained word vector dictionary is employed to encode labels of visual features, thereby enhancing the professionalism of the generated reports. Finally, a memory driven module is introduced in the decoder, addressing long-distance dependencies in medical image data. This study is validated on the chest X-ray dataset collected at Indiana University (IU X-Ray) and the medical information mart for intensive care chest x-ray (MIMIC-CXR) released by the Massachusetts Institute of Technology and Massachusetts General Hospital. Experimental results indicate that the proposed method can better focus on the affected areas, improve the accuracy and fluency of report generation, and assist radiologists in quickly completing medical image report writing.

CT radiomics-based model for predicting TMB and immunotherapy response in non-small cell lung cancer

Background Tumor mutational burden (TMB) is one of the most significant predictive biomarkers of immunotherapy efficacy in non-small cell lung cancer (NSCLC). Radiomics allows high-throughput extraction and analysis of advanced and quantitative medical imaging features. This study develops and validates a radiomic model for predicting TMB level and the response to immunotherapy based on CT features in NSCLC.Method Pre-operative chest CT images of 127 patients with NSCLC were retrospectively studied. The 3D-Slicer software was used to outline the region of interest and extract features from the CT images. Radiomics prediction model was constructed by LASSO and multiple logistic regression in a training dataset. The model was validated by receiver operating characteristic (ROC) curves and calibration curves using external datasets. Decision curve analysis was used to assess the value of the model for clinical application.ResultsA total of 1037 radiomic features were extracted from the CT images of NSCLC patients from TCGA. LASSO regression selected three radiomics features (Flatness, Autocorrelation and Minimum), which were associated with TMB level in NSCLC. A TMB prediction model consisting of 3 radiomic features was constructed by multiple logistic regression. The area under the curve (AUC) value in the TCGA training dataset was 0.816 (95% CI: 0.7109–0.9203) for predicting TMB level in NSCLC. The AUC value in external validation dataset I was 0.775 (95% CI: 0.5528–0.9972) for predicting TMB level in NSCLC, and the AUC value in external validation dataset II was 0.762 (95% CI: 0.5669–0.9569) for predicting the efficacy of immunotherapy in NSCLC.ConclusionThe model based on CT radiomic features helps to achieve cost effective improvement in TMB classification and precise immunotherapy treatment of NSCLC patients.

A Lightweight convolutional medical segmentation algorithm based on ConvNeXt to improve UNet

In recent years, UNet and its derivative networks have gained widespread recognition as major methods of medical image segmentation. However, networks like UNet often struggle with Point-of-Care (POC) healthcare applications due to their high number of parameters and computational complexity. To tackle these challenges, this paper introduces an efficient network designed for medical image segmentation called MCU-Net, which leverages ConvNeXt to enhance UNet. 1) Based on ConvNeXt, MCU-Net proposes the MCU Block, which employs techniques such as large kernel convolution, depth-wise separable convolution, and an inverted bottleneck design. To ensure stable segmentation performance, it also integrates global response normalization (GRN) layers and Gaussian Error Linear Unit (GELU) activation functions. 2) Additionally, MCU-Net introduces an enhanced Multi-Scale Convolution Attention (MSCA) module after the original UNet’s skip connections, emphasizing medical image features and capturing semantic insights across multiple scales. 3)The downsampling process replaces pooling layers with convolutions, and both upsampling and downsampling stages incorporate batch normalization (BN) layers to enhance model stability during training. The experimental results demonstrate that MCU-Net, with a parameter count of 2.19 million and computational complexity of 19.73 FLOPs, outperforms other segmentation models. The overall performance of MCU-Net in medical image segmentation surpasses that of other models, achieving a Dice score of 91.8% and mIoU of 84.7% on the GlaS dataset. When compared to UNet on the BUSI dataset, MCU-Net shows an improvement of 2% in Dice and 2.9% in mIoU.

TumorDet: A Breast Tumor Detection Model Based on Transfer Learning and ShuffleNet

Background: Breast tumor is among the most malignant tumors and early detection can improve patient’s survival rate. Currently, mammography is the most reliable method for diagnosing breast tumor because of high image resolution. Because of the rapid development of medical and artificial intelligence techniques, computer-aided diagnosis technology can greatly improve the detection accuracy of breast tumors and medical imaging has begun to use deep-learning-based approaches. In this study, the TumorDet model is proposed to detect the benign and malignant lesions of breast tumor, which has positive significance for assisting doctors in diagnosis. Objective: We use the proposed TumorDet to analyze and predict breast tumors on the real MRI dataset. Methods: (1) We introduce an adaptive gamma correction (AGC) method to balance brightness equalization and increase the contrast of mammography images; (2) we use the ShuffleNet model to exchange information between different feature layers and extract the hidden high-level features of medical images; and (3) we use the transfer learning method to fine-tune the ShuffleNet model and obtain the optimal parameters. Results: The proposed TumorDet model has shown that accuracy, sensitivity, and specificity reach 90.43%, 89.37%, and 87.81%, respectively. TumorDet performs well in the breast tumor detection task. In addition, we use the proposed TumorDet to conduct experiments on other tasks, such as forest fires, and the robustness of TumorDet is proved by experimental results. Conclusion: TumorDet employs the ShuffleNet model to exchange information between different feature layers without increasing the number of network parameters and applies transfer learning method to further extract the basic features of medical images by fine-tuning. The model is beneficial for the localization and classification of breast tumors and also performs well in forest fire detection.