Lesion Boundary Research Articles

Comprehensive Overview of the Diagnosis and Clinical Advancements in Malignant Melanoma

Malignant melanoma is the primary cause of mortality related to skin conditions and ranks as the fifth most prevalent form of cancer. The purpose of this study is to provide a comprehensive overview of the diagnosis and clinical advancements in malignant melanoma. An examination conducted under adequate lighting conditions can reveal a range of colors, such as red, white, black, and blue. The boundaries of lesions typically exhibit irregularities. The thickness of the tumor is the most crucial determinant of prognosis. Primary malignant melanoma can be categorized into different clinicohistological types, such as lentigo maligna melanoma, superficially spreading malignant melanoma, nodular malignant melanoma, acral-lentiginous melanoma, eye melanoma, and malignant melanoma of the mucous membranes. Determining oncogenic mutations in individuals with advanced melanoma can guide the use of targeted treatment, typically directed towards specific BRAF mutations.

Utility of sodium fluorescein in recurrent cervical vagus schwannoma surgery.

Cervical schwannoma is a rare neoplasm that usually occurs like a nondolent lateral neck mass but when growing and symptomatic requires radical excision. Sodium fluorescein (SF) is a dye that is uptake by schwannomas, which makes it amenable for its use in the resection of difficult or recurrent cases. We describe the case of a patient presenting with a recurrence of a vagus nerve schwannoma in the cervical region and the step-by-step technique for its complete microsurgical exeresis helped by the use of SF dye. We achieved a complete microsurgical exeresis, despite the presence of exuberant perilesional fibrosis, by exploiting the ability of SF to stain the schwannoma and nearby tissues. That happens due to altered vascular permeability, allowing us to better differentiate the lesion boundaries and reactive scar tissue under microscope visualization (YELLOW 560 nm filter). Recurrent cervical schwannoma might represent a surgical challenge due to its relation to the nerve, main cervical vessels, and the scar tissue encompassing the lesion. Although SF can cross both blood-brain and blood-tumor barriers, the impregnation of neoplastic tissue is still greater than that of nonneoplastic peripheric tissues. Such behavior may facilitate a safer removal of this kind of lesion while respecting contiguous anatomical structures.

TransRender: a transformer-based boundary rendering segmentation network for stroke lesions.

Vision transformer architectures attract widespread interest due to their robust representation capabilities of global features. Transformer-based methods as the encoder achieve superior performance compared to convolutional neural networks and other popular networks in many segmentation tasks for medical images. Due to the complex structure of the brain and the approximate grayscale of healthy tissue and lesions, lesion segmentation suffers from over-smooth boundaries or inaccurate segmentation. Existing methods, including the transformer, utilize stacked convolutional layers as the decoder to uniformly treat each pixel as a grid, which is convenient for feature computation. However, they often neglect the high-frequency features of the boundary and focus excessively on the region features. We propose an effective method for lesion boundary rendering called TransRender, which adaptively selects a series of important points to compute the boundary features in a point-based rendering way. The transformer-based method is selected to capture global information during the encoding stage. Several renders efficiently map the encoded features of different levels to the original spatial resolution by combining global and local features. Furthermore, the point-based function is employed to supervise the render module generating points, so that TransRender can continuously refine the uncertainty region. We conducted substantial experiments on different stroke lesion segmentation datasets to prove the efficiency of TransRender. Several evaluation metrics illustrate that our method can automatically segment the stroke lesion with relatively high accuracy and low calculation complexity.

Attention-based dual-path feature fusion network for automatic skin lesion segmentation

Automatic segmentation of skin lesions is a critical step in Computer Aided Diagnosis (CAD) of melanoma. However, due to the blurring of the lesion boundary, uneven color distribution, and low image contrast, resulting in poor segmentation result. Aiming at the problem of difficult segmentation of skin lesions, this paper proposes an Attention-based Dual-path Feature Fusion Network (ADFFNet) for automatic skin lesion segmentation. Firstly, in the spatial path, a Boundary Refinement (BR) module is designed for the output of low-level features to filter out irrelevant background information and retain more boundary details of the lesion area. Secondly, in the context path, a Multi-scale Feature Selection (MFS) module is constructed for high-level feature output to capture multi-scale context information and use the attention mechanism to filter out redundant semantic information. Finally, we design a Dual-path Feature Fusion (DFF) module, which uses high-level global attention information to guide the step-by-step fusion of high-level semantic features and low-level detail features, which is beneficial to restore image detail information and further improve the pixel-level segmentation accuracy of skin lesion. In the experiment, the ISIC 2018 and PH2 datasets are employed to evaluate the effectiveness of the proposed method. It achieves a performance of 0.890/ 0.925 and 0.933 /0.954 on the F1-score and SE index, respectively. Comparative analysis with state-of-the-art segmentation methods reveals that the ADFFNet algorithm exhibits superior segmentation performance.

Objective Boundary Generation for Gross Target Volume and Organs at Risk Using 3D Multi-Modal Medical Images.

Accurate delineation of Gross Target Volume (GTV) and Organs at Risk (OARs) in medical images is an essential but challenging step in radiotherapy. Deep-learning based automated delineation methods, which learn from manual annotations, are currently prevalent in academic research. However, the limited resolution of medical images and the fuzzy boundaries of lesions and organs present a challenge to the precision of manual annotations. By leveraging the complementary information from multi-modal medical images, this study proposed a novel method to generate objective boundaries of GTV and OARs. We present a novel method of objective boundary generation, inspired by image matting primarily used for 2D RGB natural images, to process 3D grayscale medical images. The proposed method has the following advantages. 1) It allows for flexible input modalities and assigns weights to each modality according to their relative significance when computing information flows in the matting algorithm. 2) It computes 3D spatial information flow among voxels, which has more advantages over its 2D counterpart. 3) It has a closed-form solution that generates deterministic results. To evaluate the characteristics of the generated boundaries, patients with stage I nasopharyngeal carcinoma (NPC) were studied, with CT images and multi-modal MR images (T1, T1C, T2) aligned using deformable registration. Region of Interests (ROIs), i.e., GTV and parotid gland, were used, with a rough trimap marking extremely few foreground voxels, many background voxels, and a large unknown region. The proposed algorithm leverages the connection between each voxel and its nearest neighbors in the feature space, to propagate the opacity information. We evaluated the results by employing both qualitative and quantitative methods. Using qualitative evaluation, experienced clinicians confirmed that the results were in agreement with the input data, especially for areas where borders were visible in most modalities (e.g., between air and tumor). For more challenging regions, where boundaries were unclear in the images, the results displayed fine-grained opacity transitions indicating the confidence of each voxel belonging to the ROI. When compared to the delineations made by clinicians, we found our results are usually more compact. We define a precision metric that evaluates the ratio of the matted foreground inside clinicians' delineations versus the entire matted foreground. Using a threshold of 0.4, our binarized result scored 0.95 for GTV and 0.92 for parotid gland. The proposed method demonstrated satisfactory results on challenging ROIs. The objective boundaries generated by this method have advantages in many aspects, including improvement of delineation protocols, enhancement of manual annotation consistency, and increase of deep-learning based automated delineation accuracy.

SKIN CANCER IMAGE SEGMENTATION USING DEEP LEARNING

Because skin cancer is a frequent and possibly lethal illness, early identification is crucial for surgical treatment. Personal dermatoscopy is crucial in the perception of skin cancer. Recent research on deep literacy techniques has shown astonishing success in automating the segmentation of skin lesions, assisting in the early diagnosis and treatment of skin cancer. This goal provides a summary of improvements in skin cancer image segmentation that are based on deep literacy. The goal of this project is to create a dependable and effective deep literacy frame for segmenting skin lesions from the dermoscopic image collection. Convolutional neural networks (CNNs), which are deep learning models that are incredibly effective at detecting daedal patterns in images, are used in the proposed model. The training data consists of a distinct collection of annotated dermoscopy images that contains nonidentical forms of skin cancer such as carcinoma, rudimentary cell melanoma, and scaled cell melanoma. To precisely describe the size of the skin abnormality, segmentation entails identifying the lesion boundaries. Transfer literacy is practiced, which entails using pre-trained convolutional neural network infrastructures, to improve the model's interpretation. Transfer literacy improves segmentation interpretation by using features discovered from large datasets. Key Words: CNN, Skin cancer, Dermoscopic, Image segmentation, Network

Design of a Novel Skin Lesion Predictor Model Using Hybrid Particle Swarm Optimization and Convolutional Neural Networks

The prediction of skin cancer at the earlier stage is extremely essential for melanoma. There is a need for intellectual computer analysis for skin lesions. The segmentation of lesion boundaries is vital to accurately identify the lesions from the dermoscopic images where the diagnosis is complex for various skin lesion types. Thus, some pre-processing steps are required to attain higher sensitive lesion boundary segmentation and classification. Initially, pre-processing is done with median filter to offer reputation of boundary preservation and does not in-cooperate newer pixel values to the processed image. Next is the process of image segmentation for Region of Interest (ROI) and non-Region of Interest ([Formula: see text] using the Jaccard Distance segmentation process. The novelty of the work relies on the inclusion of compression techniques to make the access easier without any loss. The extracted regions are encoded with Freeman Chain Coding and compressed with Lempel–Ziv–Welch (LZW) and Zero Tree Wavelet (EZW) for ROI and non-Region of Interest regions. Finally, image classification is done with Hybrid Particle Swarm Optimization and Convolutional Neural Networks ([Formula: see text]PSO-CNN). The simulation is done with TensorFlow & Python environment and the proposed model outperforms the existing standard approaches. Some metrics like objective function, confusion matrix, accuracy, precision, [Formula: see text]-measure, and recall are evaluated. The model attains 77.5% accuracy, 86.36% precision, 77.5% recall and 77.92% [Formula: see text]-measure for proposed [Formula: see text]PSO-CNN which is higher than the standard CNN model.

Own experience of using holographic neuronavigation using an augmented reality helmet in neurosurgery

Objective: to optimize surgical access to intracranial lesions (tumors, arteriovenous malformations, cysts, etc.) by using a holographic neuronavigation system with augmented reality helmet. Materials and methods: The study included thirty-four patients who underwent cerebral neurosurgical interventions. Creation and clinical use of mixed reality neuronavigation (MRN) system holograms was possible in all cases, which allowed accurate localization of lesions. The additional time required for synchronizing the MRN system with the clinical environment was estimated, which decreased with the number of MRN system uses. Operators evaluated the effectiveness of the technology and in most cases provided positive evaluations after use. Results: A semi-automatic MRN recording system on HoloLens smart glasses has been developed and tested for accuracy and performance. 34 patients with intracranial lesions were prospectively included in the study. Three-dimensional holograms of lesions, markers, and surrounding anatomical landmarks based on multimodal imaging were generated for each patient and then imported into the MRN system in AR helmet. After point recording, holograms were projected onto the patient’s head and observed through the AR helmet during the planning and conducting of surgical interventions. Conclusions: In this small pilot study, the authors found that mixed reality neuronavigation system MRN can be applied in the workflow of a neurosurgical operating room and is a possible method of preoperative identification of lesion boundaries for surgical access planning. Future studies are needed to identify strategies to improve and optimize the accuracy of MRN system.

A despeckling method for ultrasound images utilizing content-aware prior and attention-driven techniques

The despeckling of ultrasound images contributes to the enhancement of image quality and facilitates precise treatment of conditions such as tumor cancers. However, the use of existing methods for eliminating speckle noise can cause the loss of image texture features, impacting clinical judgment. Thus, maintaining clear lesion boundaries while eliminating speckle noise is a challenging task. This paper presents an innovative approach for denoising ultrasound images using a novel noise reduction network model called content-aware prior and attention-driven (CAPAD). The model employs a neural network to automatically capture the hidden prior features in ultrasound images to guide denoising and embeds the denoiser into the optimization module to simultaneously optimize parameters and noise. Moreover, this model incorporates a content-aware attention module and a loss function that preserves the structural characteristics of the image. These additions enhance the network’s capacity to capture and retain valuable information. Extensive qualitative evaluation and quantitative analysis performed on a comprehensive dataset provide compelling evidence of the model’s superior denoising capabilities. It excels in noise suppression while successfully preserving the underlying structures within the ultrasound images. Compared to other denoising algorithms, it demonstrates an improvement of approximately 5.88% in PSNR and approximately 3.61% in SSIM. Furthermore, using CAPAD as a preprocessing step for breast tumor segmentation in ultrasound images can greatly improve the accuracy of image segmentation. The experimental results indicate that the utilization of CAPAD leads to a notable enhancement of 10.43% in the AUPRC for breast cancer tumor segmentation.

Skin lesion image segmentation by using backchannel filling CNN and level sets

Pigmented skin lesion image segmentation is a key step in the computer-aided diagnosis of melanoma. This image segmentation task is also difficult due to blurred skin lesion boundaries, uneven color, and diverse textures. To improve the accuracy of skin lesion image segmentation, this paper proposes an image segmentation method that combines convolutional neural networks (CNN) and level sets. First, a backchannel filling (BCF) method is proposed. Through the first forward propagation of the CNN, the target position estimated by the CNN is backfilled into the position prior channel. Then, end-to-end deep learning models based on BCF multibranch encoding–decoding CNN (BCF-CNN) and level sets are built. The initial level set and driving energy are predicted through the second forward propagation of the BCF-CNN and then input into the level set segmentation model to complete the evolution. Additionally, a learning algorithm that combines a joint BCF-CNN and a level set is designed to train the segmentation model. The experimental results of the proposed algorithm are compared with those of other state-of-the-art algorithms on both the International Skin Imaging Collaboration (ISIC) 2017 and 2016 skin lesion challenge datasets, demonstrating that the proposed algorithm outperforms the others in terms of accuracy with a Jaccard index of approximately 77.0% and 85.1% on the ISIC 2017 and 2016 skin lesion challenge datasets, respectively.

MSREA-Net: An Efficient Skin Disease Segmentation Method Based on Multi-Level Resolution Receptive Field

Aiming at the low contrast of skin lesion image and inaccurate segmentation of lesion boundary, a skin lesion segmentation method based on multi-level split receptive field and attention is proposed. Firstly, the depth feature extraction module and multi-level splitting receptive field module are used to extract image feature information; secondly, the hybrid pooling module is used to build long-term and short-term dependencies and integrate global information and local information. Finally, the reverse residual external attention module is introduced to construct the decoding part, which can mine the potential relationship between data sets and improve the network segmentation ability. Experiments on ISBI2017 and ISIC2018 data sets show that the Dice similarity coefficient and Jaccard index reach 88.67% and 91.84%, 79.25% and 81.48%, respectively, and the accuracy reaches 93.89% and 96.16%. The segmentation method is superior to the existing algorithms as a whole. Simulation experiments show that the network has a good effect on skin lesion image segmentation and provides a new method for skin disease diagnosis.

A dual-stage transformer and MLP-based network for breast ultrasound image segmentation

Automatic segmentation of breast lesions from ultrasound images plays an important role in computer-aided breast cancer diagnosis. Many deep learning methods based on convolutional neural networks (CNNs) have been proposed for breast ultrasound image segmentation. However, breast ultrasound image segmentation is still challenging due to ambiguous lesion boundaries. We propose a novel dual-stage framework based on Transformer and Multi-layer perceptron (MLP) for the segmentation of breast lesions. We combine the Swin Transformer block with an efficient pyramid squeezed attention block in a parallel design and introduce bi-directional interactions across branches, which can efficiently extract multi-scale long-range dependencies to improve the segmentation performance and robustness of the model. Furthermore, we introduce tokenized MLP block in the MLP stage to extract global contextual information while retaining fine-grained information to segment more complex breast lesions. We have conducted extensive experiments with state-of-the-art methods on three breast ultrasound datasets, including BUSI, BUL, and MT_BUS datasets. The dice coefficient reached 0.8127 ± 0.2178, and the intersection over union reached 0.7269 ± 0.2370 on benign lesions when the Hausdorff distance was maintained at 3.75 ± 1.83. The dice coefficient of malignant lesions is improved by 3.09% for BUSI dataset. The segmentation results on the BUL and MT_BUS datasets also show that our proposed model achieves better segmentation results than other methods. Moreover, the external experiments indicate that the proposed model provides better generalization capability for breast lesion segmentation. The dual-stage scheme and the proposed Transformer module achieve the fine-grained local information and long-range dependencies to relieve the burden of radiologists.

DESIGN AND ANALYSIS OF IODINE STAINING AND NBI IN THE DIAGNOSTIC MODEL AND DIAGNOSIS OF EARLY ESOPHAGEAL CANCER AND ITS PRECANCEROUS LESIONS

Objective: This study aimed to compare the differences between narrowband imaging (NBI) and Lugo’s iodine staining endoscopy (LCE) for detecting and elucidating the site, clarity, and diagnostic accuracy of early esophageal cancer and precancerous lesions. Methods: We included patients with a high risk of developing esophageal cancer who visited the Zhongjiang County People’s Hospital between October 2020 and October 2022. Endoscopic examination was performed on each study participant. White-light endoscopy was used to observe and locate the diseased mucosa, after which NBI mode and LCE staining were used to observe the boundary between the diseased and normal mucosa. Abnormal lesions were found, and biopsies were performed on the identifiable diseased parts for pathological examination. Inflammation, LGIN, HGIN, and early esophageal cancer were diagnosed, and the differences between them were compared in terms of the two enhanced-image endoscopy techniques. A semantic segmentation model based on deep learning was adopted to assist in the diagnosis of early esophageal cancer and accurately locate cancerous areas. In order to improve its accurate diagnostic rate, we also built a semantic segmentation network model to assist in the computer-aided diagnosis of early esophageal cancer. Results: A total of 69 cases of early esophageal cancer were included in this study. The patients were aged 40–75 years, with an average age of [Formula: see text] years. Most early esophageal cancer lesions were located in the middle of the esophagus in 45/69 (65.2%) cases. LCE obtained higher clarity of lesion boundaries than NBI (59.4% versus 45.0% and 27.7% versus 15.9%, respectively), and the proportion of unclear lesions was lower (8.6% versus 21.7% and 4.3% versus 17.4%, respectively; [Formula: see text]). For early esophageal cancer and precancerous lesions, the missed-diagnosis rate of white-light endoscopy was 20%, that of NBI was 3.15%, and that of LCE was 0%, with statistical significance ([Formula: see text]). The rate of missed diagnosis of LGIN (three cases) by NBI was 3.15%, which was not significantly different from that of LCE ([Formula: see text]). We found that the lesion area could be more accurately determined using deep learning models to segment NBI images. By constructing a deep learning model for the diagnosis and classification of esophageal cancer, its diagnostic rate rose to 99.5%. Conclusion: The age range of patients with early esophageal cancer in Zhongjiang County, Sichuan Province, was 40–75 years, and tumors mostly occurred in the middle of the esophagus. The boundary clarity of early esophageal carcinomas was higher on LCE than that on NBI. The diagnostic accuracies of NBI and LCE are much higher than those of conventional white-light endoscopy. Therefore, both LCE and NBI are helpful for detecting early esophageal cancer and precancerous lesions. Moreover, the diagnostic rate of clinical esophageal cancer can be effectively improved using a dedicated deep learning network model.

First Report of Olive Mild Mosaic Virus in Imported Tulips (Tulipa gesneriana) in Korea.

Tulip cultivation in Korea primarily uses imported bulbs due to the absence of domestic production. To ensure safety and sustainability, Korean authorities have implemented strict phytosanitary measures for five viruses: arabis mosaic virus, tobacco necrosis virus, tobacco ringspot virus, tomato black ring virus, and tomato bushy stunt virus. In April 2021, 86 tulip plants presented symptoms such as chlorotic mottle, mosaic, streak, stripe, yellowing of the leaves, and color breaking on flowers. These samples were collected to investigate the incidence of viruses in four Korean provinces (Gangwon, Gyeongbuk, Gyeongnam, and Chungnam). The leaves and petals from each individual sample (10 mg each) were pooled and ground using liquid nitrogen. Total RNA was extracted using a Maxwell® 16 LEV Plant RNA Kit (Promega, Madison, USA). A cDNA library was constructed using TruSeq Standard Total RNA with Ribo - Zero (Illumina, San Diego, USA) and sequenced on an Illumina NovaSeq 6000 platform (Macrogen, Seoul, Korea) with 100-bp paired-end reads. Trinity software identified tulip breaking virus (TBV), tulip virus X (TVX), and lily symptomless virus (LSV), which are known to occur in Korea (Bak et al. 2023) by de novo assembly of 628 million reads into 498,795 contigs. The contigs were annotated as previously described (Bak et al. 2022). Moreover, a contig (ON758350) related to olive mild mosaic virus (OMMV; genus Alphanecrovirus, family Tombusviridae) was identified through BLASTn analysis. This contig had a 99.27% nucleotide (nt) identity to OMMV PPO-L190209 (KU641010), which was assembled from 201,346 reads and spanned 3,713 bp. To confirm the presence of OMMV, a primer pair (5'-GAATGTCTGGCGTTAAGCG-3'/5'-GTGTCCTGCGCATCATACAC-3') was designed to amplify a 797-bp fragment of the coat protein gene. In RT-PCR, 31.4% (27/86) of samples were positive for OMMV and coinfected with TBV or TBV+LSV. Coinfection with TBV led to chlorotic mottling and stripes, whereas triple coinfection with TBV+LSV produced distinct yellow streaks and mosaic within the lesion boundaries. In contrast, solely TBV infection did not produce such symptoms. The samples infected with OMMV were exclusively collected from Gangwon and Gyeongnam. In each province, an RT-PCR amplicon was cloned, and subsequently sequenced (Bioneer, Daejeon, Korea). The obtained sequences were named CC (OM243091) and GS (OM243092), and they shared 98.6% and 98.9% identity with PPO-L190209 (KU641010), respectively. A bioassay was conducted using a leaf infected with OMMV CC and TBV to inoculate 13 indicator species in triplicate, including Capsicum annuum, Chenopodium amaranticolor, C. quinoa, Cucumis sativus, Nicotiana benthamiana, N. clevelandii, N. glutinosa, N. occidentalis, N. rustica, N. tabacum, Solanum lycopersicum, Tetragonia tetragonioides, and Tulipa gesneriana. The RT-PCR revealed positivity only for OMMV in the upper leaves of N. clevelandii, while all other species were negative with no symptoms. To our knowledge, this is the first report of OMMV occurring in tulips grown from imported bulbs in Korea, with no other known natural hosts such as olive tree (Cardoso et al. 2004), spinach (Gratsia et al. 2012), and corn salad (Verdin et al. 2018). The Korean OMMV isolates exhibited a high nt identity with the foreign isolate, and the samples were collected from farms that rely entirely on bulb imports for cultivation. These suggest that the outbreak of OMMV was likely caused by imported bulbs.

A novel endoscopy image fusion system: combine white light imaging and compound band imaging.

White light imaging (WLI) is a commonly seen examination mode in endoscopy. The particular light in compound band imaging (CBI) can highlight delicate structures, such as capillaries and tiny structures on the mucosal surface. These two modes complement each other, and doctors switch between them manually to complete the examination. This paper proposes an endoscopy image fusion system to combine WLI and CBI. We add a real-time rotatable color wheel in the light source device of the AQ-200 endoscopy system to achieve rapid imaging of two modes at the same position of living tissue. The two images corresponding to the pixel level can avoid registration and lay the foundation for image fusion. We propose a multi-scale image fusion framework, which involves Laplacian pyramid (LP) and convolutional sparse representation (CSR) and strengthens the details in the fusion rule. Volunteer experiments and ex vivo pig stomach trials are conducted to verify the feasibility of our proposed system. We also conduct comparative experiments with other image fusion methods, evaluate the quality of the fused images, and verify the effectiveness of our fusion framework. The results show that our fused image has rich details, high color contrast, apparent structures, and clear lesion boundaries. An endoscopy image fusion system is proposed, which does not change the doctor's operation and makes the fusion of WLI and CBI optical staining technology a reality. We change the light source device of the endoscope, propose an image fusion framework, and verify the feasibility and effectiveness of our scheme. Our method fully integrates the advantages of WLI and CBI, which can help doctors make more accurate judgments than before. The endoscopy image fusion system is of great significance for improving the detection rate of early lesions and has broad application prospects.

Generative adversarial networks based skin lesion segmentation

Skin cancer is a serious condition that requires accurate diagnosis and treatment. One way to assist clinicians in this task is using computer-aided diagnosis tools that automatically segment skin lesions from dermoscopic images. We propose a novel adversarial learning-based framework called Efficient-GAN (EGAN) that uses an unsupervised generative network to generate accurate lesion masks. It consists of a generator module with a top-down squeeze excitation-based compound scaled path, an asymmetric lateral connection-based bottom-up path, and a discriminator module that distinguishes between original and synthetic masks. A morphology-based smoothing loss is also implemented to encourage the network to create smooth semantic boundaries of lesions. The framework is evaluated on the International Skin Imaging Collaboration Lesion Dataset. It outperforms the current state-of-the-art skin lesion segmentation approaches with a Dice coefficient, Jaccard similarity, and accuracy of 90.1%, 83.6%, and 94.5%, respectively. We also design a lightweight segmentation framework called Mobile-GAN (MGAN) that achieves comparable performance as EGAN but with an order of magnitude lower number of training parameters, thus resulting in faster inference times for low compute resource settings.

A Fuzzy Transformer Fusion Network (FuzzyTransNet) for Medical Image Segmentation: The Case of Rectal Polyps and Skin Lesions

Skin melanoma, one of the deadliest forms of cancer worldwide, demands precise diagnosis to mitigate cancer-related mortality. While histopathological examination, characterized by its cost-effectiveness and efficiency, remains the primary diagnostic approach, the development of an accurate detection system is pressing due to melanoma’s varying sizes, shapes, and indistinct boundaries shared with normal tissues. To address the efficient segmentation of skin melanoma, we propose an innovative hybrid neural network approach in this study. Initially, a fuzzy neural network is constructed using fuzzy logic to preprocess medical images, supplemented by wavelet transformation for image enhancement. Subsequently, the Swin Transformer V2 and ResNet50 networks are introduced to parallelly extract features and apply them to the task of skin melanoma segmentation. Extensive experimental comparisons are conducted with other classic and advanced medical segmentation algorithms on publicly available skin datasets, namely ISIC 2017 and ISIC 2018. Experimental results reveal that our method outperforms the optimal algorithms by 1.3% in the Dice coefficient and 1.3% in accuracy on the ISIC 2018 dataset. The evaluation metrics indicate the effectiveness of the constructed fuzzy block in identifying uncertain lesion boundaries, while the Transformer–CNN branch adeptly extracts global features while accurately capturing underlying details. Additionally, we successfully apply our method to colon polyp segmentation tasks with similar indistinct boundaries, achieving remarkable segmentation outcomes.

CFHA-Net: A polyp segmentation method with cross-scale fusion strategy and hybrid attention

Colorectal cancer is a prevalent disease in modern times, with most cases being caused by polyps. Therefore, the segmentation of polyps has garnered significant attention in the field of medical image segmentation. In recent years, the variant network derived from the U-Net network has demonstrated a good segmentation effect on polyp segmentation challenges. In this paper, a polyp segmentation model, called CFHA-Net, is proposed, that combines a cross-scale feature fusion strategy and a hybrid attention mechanism. Inspired by feature learning, the encoder unit incorporates a cross-scale context fusion (CCF) module that performs cross-layer feature fusion and enhances the feature information of different scales. The skip connection is optimized by proposed triple hybrid attention (THA) module that aggregates spatial and channel attention features from three directions to improve the long-range dependence between features and help identify subsequent polyp lesion boundaries. Additionally, a dense-receptive feature fusion (DFF) module, which combines dense connections and multi-receptive field fusion modules, is added at the bottleneck layer to capture more comprehensive context information. Furthermore, a hybrid pooling (HP) module and a hybrid upsampling (HU) module are proposed to help the segmentation network acquire more contextual features. A series of experiments have been conducted on three typical datasets for polyp segmentation (CVC-ClinicDB, Kvasir-SEG, EndoTect) to evaluate the effectiveness and generalization of the proposed CFHA-Net. The experimental results demonstrate the validity and generalization of the proposed method, with many performance metrics surpassing those of related advanced segmentation networks. Therefore, proposed CFHA-Net could present a promising solution to the challenges of polyp segmentation in medical image analysis. The source code of proposed CFHA-Net is available at https://github.com/CXzhai/CFHA-Net.git.

Image segmentation and feature extraction method for lung lesion detection in computed tomography images

Lung cancer is a form of cancer that causes uncontrollable cell growth in the lungs. Patients with lung cancer frequently miss a treatment, face higher health care costs, and get the worst outcomes. The detection of the existence of lung cancer can be performed in a variety of ways, such as computed tomography (CT), magnetic resonance imaging (MRI), and radiography. Many researchers have developed ways of automating lung cancer diagnosis using image processing techniques because of the noise and low image quality between the cancer cells, the lung, and the background. This study develops an image processing technique that uses image segmentation algorithms to segment lung nodules in computed tomography images using feature extraction. In the initial phase, it is essential to establish a rigorous image processing framework with the following sequential steps: (i) object edge identification and (ii) lesion boundary recognition. The architecture includes image processing techniques, thresholding, and morphological detections (erosion and dilation). Lesions can have various sizes and shapes, both regular and irregular. The new method has been applied to find the lesions using their roundness size. In addition to learning purely from CT scans, the previously studied lesion characteristics are also integrated. Data was collected from the Advanced Medical and Dental Institute (AMDI), Universiti Sains Malaysia, Penang. The manual segmentation was used image segmented in the MATLAB software function to remove the background of the images. The perimeter evaluates such as accuracy, recall, and F-score. Based on the analysis the performance of lung lesion segmentation of accuracy is 99.95, recall at 45.76%, and the F-score is 60.67%. For lung lesion detection, the results shows it consist of 3-5 slices with the value of roundness. Besides, lesion detection also have continuity for the roundness value. The experiment results found clear support for the next step of this research for classifications of lesions.

CKD-TransBTS: Clinical Knowledge-Driven Hybrid Transformer with Modality-Correlated Cross-Attention for Brain Tumor Segmentation.

Brain tumor segmentation (BTS) in magnetic resonance image (MRI) is crucial for brain tumor diagnosis, cancer management and research purposes. With the great success of the ten-year BraTS challenges as well as the advances of CNN and Transformer algorithms, a lot of outstanding BTS models have been proposed to tackle the difficulties of BTS in different technical aspects. However, existing studies hardly consider how to fuse the multi-modality images in a reasonable manner. In this paper, we leverage the clinical knowledge of how radiologists diagnose brain tumors from multiple MRI modalities and propose a clinical knowledge-driven brain tumor segmentation model, called CKD-TransBTS. Instead of directly concatenating all the modalities, we re-organize the input modalities by separating them into two groups according to the imaging principle of MRI. A dual-branch hybrid encoder with the proposed modality-correlated cross-attention block (MCCA) is designed to extract the multi-modality image features. The proposed model inherits the strengths from both Transformer and CNN with the local feature representation ability for precise lesion boundaries and long-range feature extraction for 3D volumetric images. To bridge the gap between Transformer and CNN features, we propose a Trans&CNN Feature Calibration block (TCFC) in the decoder. We compare the proposed model with six CNN-based models and six transformer-based models on the BraTS 2021 challenge dataset. Extensive experiments demonstrate that the proposed model achieves state-of-the-art brain tumor segmentation performance compared with all the competitors.