Lesions In Region Research Articles

Modeling Functional Brain Networks for ADHD via Spatial Preservation-Based Neural Architecture Search.

Modeling functional brain networks (FBNs) for attention deficit hyperactivity disorder (ADHD) has sparked significant interest since the abnormal functional connectivity is discovered in certain functional magnetic resonance imaging (fMRI)-based brain regions compared to typical developmental control (TC) individuals. However, existing models for modeling FBNs generally use dimensionality reduction techniques to process the high dimensional input data, which results in confusion and an inaccurate representation of voxel interactions between spatially close brain regions, causing misdiagnosis of the disease. To address these issues, we propose a spatial preservation-based neural architecture search (SP-NAS) for FBNs modeling in ADHD. The main work includes three-fold: 1) A spatial preservation module is designed to embed original spatial information into dimensionality reduction data, addressing the challenge of a large number of parameters in the original data and mitigating disease misdiagnosis resulting from voxel confusion between different brain regions caused by dimensionality reduction. 2) A search space using more suitable search operations is constructed to efficiently extract spatial-temporal interaction characteristics of fMRI data in ADHD while narrowing the search space. 3) Cross-regional association differences between ADHD and TC groups are explored for ADHD auxiliary diagnosis since the abnormal activation regions of ADHD relative to TC on the brain regions and the abnormal connectivity between the lesion brain regions are identified. Model validation results on the ADHD-200 dataset show that the FBNs obtained from SP-NAS not only achieve competitive results in ADHD diagnosis but also reveal abnormal connections in the lesion regions of ADHD consistent with clinical diagnosis.

Diffusion network with spatial channel attention infusion and frequency spatial attention for brain tumor segmentation.

Accurate segmentation of gliomas is crucial for diagnosis, treatment planning, and evaluating therapeutic efficacy. Physicians typically analyze and delineate tumor regions in brain magnetic resonance imaging (MRI) images based on personal experience, which is often time-consuming and subject to individual interpretation. Despite advancements in deep learning technology for image segmentation, current techniques still face challenges in clearly defining tumor boundary contours and enhancing segmentationaccuracy. To address these issues, this paper proposes a conditional diffusion network (SF-Diff) with a spatial channel attention infusion (SCAI) module and a frequency spatial attention (FSA) mechanism to achieve accurate segmentation of the whole tumor (WT) region in braintumors. SF-Diff initially extracts multiscale information from multimodal MRI images and subsequently employs a diffusion model to restore boundaries and details, thereby enabling accurate brain tumor segmentation (BraTS). Specifically, a SCAI module is developed to capture multiscale information within and between encoder layers. A dual-channel upsampling block (DUB) is designed to assist in detail recovery during upsampling. A FSA mechanism is introduced to better match the conditional features with the diffusion probability distribution information. Furthermore, a cross-model loss function has been implemented to supervise the feature extraction of the conditional model and the noise distribution of the diffusionmodel. The dataset used in this paper is publicly available and includes 369 patient cases from the Multimodal BraTS Challenge 2020 (BraTS2020). The conducted experiments on BraTS2020 demonstrate that SF-Diff performs better than other state-of-the-art models. The method achieved a Dice score of 91.87%, a Hausdorff 95 of 5.47 mm, an IoU of 84.96%, a sensitivity of 92.29%, and a specificity of 99.95% onBraTS2020. The proposed SF-Diff performs well in identifying the WT region of the brain tumors compared to other state-of-the-art models, especially in terms of boundary contours and non-contiguous lesion regions, which is clinically significant. In the future, we will further develop this method for brain tumor three-class segmentation task.

Association Between Arch-Shaped Hypo-Autofluorescent Lesions Detected Using Fundus Autofluorescence and Postoperative Hypotony.

Background: Chorioretinal folds are observed after vitrectomy due to ocular collapse caused by low intraocular pressure. The purpose of this study is to investigate the relationship between the postoperative hypotony, chorioretinal folds, and the fundus autofluorescence (FAF) findings. Methods: Two-hundred-and-seventy consecutive eyes that had undergone 25- or 27-gauge vitrectomy were examined. The associations between the arch-shaped hypo-autofluorescent lesions in the FAF images and the postoperative hypotony with intraocular pressure (IOP) ≤ 4 mmHg were determined on the day after the surgery. Results: Arch-shaped hypo-autofluorescent lesions were seen in 4 of the 270 eyes (1.5%), and hypo-autofluorescence was observed in 3 of 14 hypotonic eyes (18.5%). This was significantly more frequent than in the non-hypotony group (0.4%, p = 0.0004). Optical coherence tomography showed a loss of the ellipsoid zone and retinal pigment epithelial layer in the region of the arch-shaped lesions. None of the arch-shaped hypo-autofluorescent lesions involved the fovea, and the vision recovered in all cases. The hypo-autofluorescent lesions did not disappear during the 4 to 16 month observation period. Conclusions: The postoperative arch-shaped hypo-autofluorescent lesions were associated with postoperative hypotony and RPE damage due to chorioretinal folds. These findings remained even when the IOP was normalized and chorioretinal folds disappeared.

A Comprehensive and Repeatable Contrast-Enhanced Ultrasound Quantification Approach for Clinical Evaluations of Tumor Blood Flow.

The aim of this study is to define a comprehensive and repeatable contrast-enhanced ultrasound (CEUS) imaging protocol and analysis method to quantitatively assess lesional blood flow. Easily repeatable CEUS evaluations are essential for longitudinal treatment monitoring. The quantification method described here aims to provide a structure for future clinical studies. This retrospective analysis study included liver CEUS studies in 80 patients, 40 of which contained lesions (primarily hepatocellular carcinoma, n = 28). Each patient was given at least 2 injections of a microbubble contrast agent, and 60-second continuous loops were acquired for each injection to enable evaluation of repeatability. For each bolus injection, 1.2 mL of contrast was delivered, whereas continuous, stationary scanning was performed. Automated respiratory gating and motion compensation algorithms dealt with breathing motion. Similar in size regions of interest were drawn around the lesion and liver parenchyma, and time-intensity curves (TICs) with linearized image data were generated. Four bolus transit parameters, rise time (RT), mean transit time (MTT), peak intensity (PI), and area under the curve (AUC), were extracted either directly from the actual TIC data or from a lognormal distribution curve fitted to the TIC. Interinjection repeatability for each parameter was evaluated with coefficient of variation. A 95% confidence interval was calculated for all fitted lognormal distribution curve coefficient of determination (R2) values, which serves as a data quality metric. One-sample t tests were performed between values obtained from injection pairs and between the fitted lognormal distribution curve and direct extraction from the TIC calculation methods to establish there were no significant differences between injections and measurement precision, respectively. Average interinjection coefficient of variation with both the fitted curve and direct calculation of RT and MTT was less than 21%, whereas PI and AUC were less than 40% for lesion and parenchyma regions of interest. The 95% confidence interval for the R2 value of all fitted lognormal curves was [0.95, 0.96]. The 1-sample t test for interinjection value difference showed no significant differences, indicating there was no relationship between the order of the repeated bolus injections and the resulting parameters. The 1-sample t test between the values from the fitted lognormal distribution curve and the direct extraction from the TIC calculation found no statistically significant differences (α = 0.05) for all perfusion-related parameters except lesion and parenchyma PI and lesion MTT. The scanning protocol and analysis method outlined and validated in this study provide easily repeatable quantitative evaluations of lesional blood flow with bolus transit parameters in CEUS data that were not available before. With vital features such as probe stabilization ideally performed with an articulated arm and an automated respiratory gating algorithm, we were able to achieve interinjection repeatability of blood flow parameters that are comparable or surpass levels currently established for clinical 2D CEUS scans. Similar values and interinjection repeatability were achieved between calculations from a fitted curve or directly from the data. This demonstrated not only the strength of the protocol to generate TICs with minimal noise, but also suggests that curve fitting might be avoided for a more standardized approach. Utilizing the imaging protocol and analysis method defined in this study, we aim for this methodology to potentially assist clinicians to assess true perfusion changes for treatment monitoring with CEUS in longitudinal studies.

Technical aspects of the new BYCROSSTM atherectomy device - preliminary results after 28 patients.

Purpose: Technical aspects are crucial for the planning and performing of the atherectomy to treat peripheral arterial disease. This report illustrates the use of a novel atherectomy device and investigates the feasibility, efficacy, and safety procedures involved in performing the atherectomy on 28 patients. We performed a prospective analysis of patients who underwent an atherectomy with the BYCROSSTM Atherectomy device between August 2022 and September 2023 at a tertiary referral centre. Patients with a lesion below the aortic bifurcation (vessel diameter ≥ 3mm) having a de novo or restenotic (stent-included) present were recruited. Major adverse events (MAE) are defined as amputation, death, myocardial infarction, or angiographic distal embolization that require a separate intervention. Of the 28 patients treated with the BYCROSS device, 23 were men with a mean age of 65.6 ± 9,6 years and a mean BMI of 24,6 ± 3.9 kg/m2. Most patients had a typical atherogenic risk profile as well as multiple preexisting comorbidities. In all patients, a symptomatic peripheral arterial disease (PAD) was the main reason for an intervention. The most common Rutherford category was 5 (12/28). The most common lesion region was the femoropopliteal segment (25/28) with 23 de novo stenosis. Mean lesion length was 218,0 ± 93,7 mm. The mean PACCS Score was 3,0 ± 1,0. Stenosis grade was by mean 99,3 ± 3,7%. Ankle Brachial Index (ABI) increased significantly after BYCROSS atherectomy (pre- 0,44 ± 0,43 vs. post-procedure 0,84 ± 0,30 P<0,001. Target lesion/vessel revascularization (TLR/TVR) within the first 30 days was 3,6% (1/28). 30-day MAE rate was 14,3% (vessel perforation in 3/28 patients, embolism in 1/28). There were no deaths, index limb amputations, or myocardial infarctions. The BYCROSSTM atherectomy system is a new device with numerous advantages in treating high-grade, calcifying stenosis and occlusion processes in PAD. Based on the above findings, the BYCROSSTM Atherectomy device represents a feasible, safe, and effective method for endovascular treatment of peripheral arterial disease.

Clinical Meaningfulness of an Algorithm-Based Service for Analyzing Treatment Response in Patients with Metastatic Cancer Using FDG PET/CT.

Background/Objectives: Determining how a patient with metastatic cancer is responding to therapy can be difficult for medical oncologists, especially with text-only radiology reports. In this investigation, we assess the clinical usefulness of a new algorithm-based analysis that provides spatial location and quantification for each detected lesion region of interest (ROI) and compare it to information included in radiology reports in the United States. Methods: Treatment response radiology reports for FDG PET/CT scans were retrospectively gathered from 228 patients with metastatic cancers. Each radiology report was assessed for the presence of both qualitative and quantitative information. A subset of patients (N = 103) was further analyzed using an algorithm-based service that provides the clinician with comprehensive quantitative information, including change over time, of all detected ROI with visualization of anatomical location. For each patient, three medical oncologists from different practices independently rated the usefulness of the additional analysis overall and in four subcategories. Results: In the 228 radiology reports, quantitative information of size and uptake was provided for at least one lesion at one time point in 78% (size) and 95% (uptake) of patients. This information was reported for both analyzed time points (current scan and previous comparator) in 52% (size) and 66% (uptake) of patients. Only 7% of reports quantified the total number of lesions, and none of the reports quantified changes in all lesions for patients with more than a few lesions. In the assessment of the augmentative algorithm-based analysis, the majority of oncologists rated it as overall useful for 98% of patients (101/103). Within specific categories of use, the majority of oncologists voted to use it for making decisions regarding systemic therapy in 97% of patients, for targeted therapy decisions in 72% of patients, for spatial location information in 96% of patients, and for patient education purposes in 93% of patients. Conclusions: For patients with metastatic cancer, the algorithm-based analysis of all ROI would allow oncologists to better understand treatment response and support their work to more precisely optimize the patient's therapy.

Skin Lesion Segmentation through Generative Adversarial Networks with Global and Local Semantic Feature Awareness

The accurate segmentation of skin lesions plays an important role in the diagnosis and treatment of skin cancers. However, skin lesion areas are rich in details and local features, including the appearance, size, shape, texture, etc., which pose challenges for the accurate localization and segmentation of the target area. Unfortunately, the consecutive pooling and stride convolutional operations in existing convolutional neural network (CNN)-based solutions lead to the loss of some spatial information and thus constrain the accuracy of lesion region segmentation. In addition, using only the traditional loss function in CNN cannot ensure that the model is adequately trained. In this study, a generative adversarial network is proposed, with global and local semantic feature awareness (GLSFA-GAN) for skin lesion segmentation based on adversarial training. Specifically, in the generator, a multi-scale localized feature fusion module and an effective channel-attention module are designed to acquire the multi-scale local detailed information of the skin lesion area. In addition, a global context extraction module in the bottleneck between the encoder and decoder of the generator is used to capture more global semantic features and spatial information about the lesion. After that, we use an adversarial training strategy to make the discriminator discern the generated labels and the segmentation prediction maps, which assists the generator in yielding more accurate segmentation maps. Our proposed model was trained and validated on three public skin lesion challenge datasets involving the ISIC2017, ISIC2018, and HAM10000, and the experimental results confirm that our proposed method provides a superior segmentation performance and outperforms several comparative methods.

Assessment of image-derived input functions from small vessels for patlak parametric imaging using total-body PET/CT.

The image-derived input function (IDIF) from the descending aorta has demonstrated performance comparable to arterial blood sampling while avoiding its invasive nature in parametric imaging. However, in conventional PET, large vessels may not always be within the imaging field of view (FOV). This study aims to evaluate the efficacy of dynamic parametric Ki imaging using image-derived input functions (IDIFs) extracted from various arteries, facilitated by total-body PET/CT. Twenty-three participants underwent a 60-minute total-body [18F]FDG PET scan. Data from each subject were used to reconstruct both total-body PET images and short-axis field-of-view PET images at different bed positions, each with a 25cm axial field-of-view (AFOV). Partial volume correction (PVC) was performed using the blurred Van Cittert iterative deconvolution. IDIFs extracted from the descending aorta, carotid artery, abdominal aorta, and iliac artery were employed for Patlak analysis. The resulting Ki images were compared using quantification indicators and subjective assessment. Linear regression analysis was conducted to examine the correlation of Ki values among IDIFs in normal organ and lesion regions of interest (ROIs). High similarities were observed in Ki images derived from the IDIFs from the descending aorta and other arteries, with a median structural similarity index measure (SSIM) above 0.98 and a median peak signal-to-noise ratio (PSNR) above 37dB. Linear regression analysis revealed strong correlations in Ki values (r² > 0.88) between the descending aorta and the three alternative vessels, with slopes of the linear fits close to 1. No significant difference in lesion detectability among IDIFs was found, as assessed visually and using metrics such as tumor-to-background ratio (TBR) and contrast-to-noise ratio (CNR) (P < 0.05). IDIFs from smaller vessels can reliably reconstruct parametric Ki images without compromising lesion detectability, providing clinically relevant information.

Quantitative Assessment of Polarization and Elastic Properties of Endometrial Tissue for Precancer/Cancer Diagnostics Using Multimodal Optical Coherence Tomography

Objectives: The most important phase in the endometrial pathologies diagnostics is the histological examination of tissue biopsies obtained under visual hysteroscopic control. However, the unclear visual diagnostics characteristics of subtle focal endometrial pathologies often lead to selection errors regarding suspicious endometrial lesions and to a subsequent false pathological diagnosis/underestimation of precancer or early-stage cancer. Methods: In this study, we investigate the potential of Multimodal Optical Coherence Tomography (MM OCT) to verify suspicious endometrial lesion regions before biopsy collection. We study the polarization (by cross-polarization OCT, CP OCT) and elastic (by compression OCT-elastography, C-OCE) properties of ex vivo endometrial tissue samples in normal conditions (proliferative and secretory phases to the menstrual cycle, atrophic endometrium) with endometrial hyperplasia (non-atypical and endometrial intraepithelial neoplasia) and endometrial cancer subtypes (low-grade, high-grade, clear cell and serous). Results: To the best of our knowledge, this is the first quantitative assessment of relevant OCT parameters (depth-resolved attenuation coefficient in co-[Att(co) values] and cross-[(Att(cross) values] polarizations and Young’s elastic modulus [stiffness values]) for the selection of the most objective criteria to identify the clinically significant endometrial pathologies: endometrial intraepithelial neoplasia and endometrial cancer. The study demonstrates the possibility of detecting endometrial pathologies and establishing optimal threshold values of MM OCT criteria for the identification of endometrial cancer using CP OCT (by Att(co) values = 3.69 mm−1, Sensitivity (Se) = 86.1%, Specificity (Sp) = 92.6%; by Att(cross) values = 2.27 mm−1, Se = 86.8%, Sp = 87.0%) and C-OCE (by stiffness values = 122 kPa, Se = 93.2%, Sp = 91.1%). The study also differentiates endometrial intraepithelial neoplasia from non-atypical endometrial hyperplasia and normal endometrium using C-OCE (by stiffness values = 95 kPa, Se = 87.2%, Sp = 90.1%). Conclusions: The results are indicative of the efficacy and potential of clinical implementation of in vivo hysteroscopic-like MM OCT in the diagnosis of endometrial pathologies.

An ensemble machine learning model assists in the diagnosis of gastric ectopic pancreas and gastric stromal tumors

ObjectiveTo develop an ensemble machine learning (eML) model using multiphase computed tomography (MPCT) for distinguishing between gastric ectopic pancreas (GEP) and gastric stromal tumors (GIST) in lesions < 3 cm.MethodsIn this study, we retrospectively collected MPCT images from 138 patients between April 2017 and June 2023 across two centers. Cohort 1 comprised 94 patients divided into a training cohort and an internal validation cohort, while the 44 patients from Cohort 2 constituted the external validation cohort. Deep learning (DL) models were constructed based on the lesion region, and radiomics features were extracted to develop radiomics models, which were later integrated into the fusion model. Model performance was assessed through the analysis of the area under the receiver operating characteristic curve (AUROC). The diagnostic efficacy of the optimal model was compared with that of a radiologist. Additionally, the radiologist with the assistance of the eML model provides a secondary diagnosis, to assess the potential clinical value of the model.ResultsAfter evaluation using an external validation cohort, the radiomics model demonstrated the highest performance in the venous phase, achieving AUROC of 0.87. The DL model showed optimal performance in the non-contrast phase, with AUROC of 0.81. The eML achieved the best performance across all models, with AUROC of 0.90. The use of eML-assisted analysis resulted in a significant improvement in the junior radiologist’s accuracy, rising from 0.77 to 0.93 (p < 0.05). However, the senior radiologist’s accuracy, while improving from 0.86 to 0.95, did not exhibit a statistically significant difference.ConclusioneML model based on MPCT can effectively distinguish between GEPs and GISTs < 3 cm.Critical relevance statementThe multiphase CT-based fusion model, incorporating radiomics and DL technology, proves effective in distinguishing between GEP and gastric stromal tumors, serving as a valuable tool to enhance diagnoses and offering references for clinical decision-making.Key PointsNo studies yet differentiated these tumors via radiomics or DL.Radiomics and DL methodologies unveil potentially distinct phenotypes within lesions.Quantitative analysis on CT for GIST and ectopic pancreas.Ensemble learning aids accurate diagnoses, assisting treatment decisions.Graphical

Discrimination Between Benign and Malignant Lesions With Restriction Spectrum Imaging MRI in an Enriched Breast Cancer Screening Cohort.

Breast cancer screening with dynamic contrast-enhanced MRI (DCE-MRI) is recommended for high-risk women but has limitations, including variable specificity and difficulty in distinguishing cancerous (CL) and high-risk benign lesions (HRBL) from average-risk benign lesions (ARBL). Complementary non-invasive imaging techniques would be useful to improve specificity. To evaluate the performance of a previously-developed breast-specific diffusion-weighted MRI (DW-MRI) model (BS-RSI3C) to improve discrimination between CL, HRBL, and ARBL in an enriched screening population. Prospective. Exactly 187 women, either with mammography screening recommending additional imaging (N = 49) or high-risk individuals undergoing routine breast MRI (N = 138), before the biopsy. Multishell DW-MRI echo planar imaging sequence with a reduced field of view at 3.0 T. A total of 72 women had at least one biopsied lesion, with 89 lesions categorized into ARBL, HRBL, CL, and combined CLs and HRBLs (CHRLs). DW-MRI data were processed to produce apparent diffusion coefficient (ADC) maps, and estimate signal contributions (C1, C2, and C3-restricted, hindered, and free diffusion, respectively) from the BS-RSI3C model. Lesion regions of interest (ROIs) were delineated on DW images based on suspicious DCE-MRI findings by two radiologists; control ROIs were drawn in the contralateral breast. One-way ANOVA and two-sided t-tests were used to assess differences in signal contributions and ADC values among groups. P-values were adjusted using the Bonferroni method for multiple testing, P = 0.05 was used for the significance level. Receiver operating characteristics (ROC) curves and intra-class correlations (ICC) were also evaluated. C1, √C1C2, and were significantly different in HRBLs compared with ARBLs (P-values < 0.05). The had the highest AUC (0.821) in differentiating CHRLs from ARBLs, performing better than ADC (0.696), especially in non-mass enhancement (0.776 vs. 0.517). This study demonstrated the BS-RSI3C could differentiate HRBLs from ARBLs in a screening population, and separate CHRLs from ARBLs better than ADC. 2.

DFEDC: Dual fusion with enhanced deformable convolution for medical image segmentation

Considering the complexity of lesion regions in medical images, current researches relying on CNNs typically employ large-kernel convolutions to expand the receptive field and enhance segmentation quality. However, these convolution methods are hindered by substantial computational requirements and limited capacity to extract contextual and multi-scale information, making it challenging to efficiently segment complex regions. To address this issue, we propose a dual fusion with enhanced deformable convolution network, namely DFEDC, which dynamically adjusts the receptive field and simultaneously integrates multi-scale feature information to effectively segment complex lesion areas and process boundaries. Firstly, we combine global channel and spatial fusion in a serial way, which integrates and reuses global channel attention and fully connected layers to achieve lightweight extraction of channel and spatial information. Additionally, we design a structured deformable convolution (SDC) that structures deformable convolution with inceptions and large kernel attention, and enhances the learning of offsets through parallel fusion to efficiently extract multi-scale feature information. To compensate for the loss of spatial information of SDC, we introduce a hybrid 2D and 3D feature extraction module to transform feature extraction from a single dimension to a fusion of 2D and 3D. Extensive experimental results on the Synapse, ACDC, and ISIC-2018 datasets demonstrate that our proposed DFEDC achieves superior results.

An attentional mechanism model for segmenting multiple lesion regions in the diabetic retina.

Diabetic retinopathy (DR), a leading cause of blindness in diabetic patients, necessitates the precise segmentation of lesions for the effective grading of lesions. DR multi-lesion segmentation faces the main concerns as follows. On the one hand, retinal lesions vary in location, shape, and size. On the other hand, the currently available multi-lesion region segmentation models are insufficient in their extraction of minute features and are prone to overlooking microaneurysms. To solve the above problems, we propose a novel deep learning method: the Multi-Scale Spatial Attention Gate (MSAG) mechanism network. The model inputs images of varying scales in order to extract a range of semantic information. Our innovative Spatial Attention Gate merges low-level spatial details with high-level semantic content, assigning hierarchical attention weights for accurate segmentation. The incorporation of the modified spatial attention gate in the inference stage enhances precision by combining prediction scales hierarchically, thereby improving segmentation accuracy without increasing the associated training costs. We conduct the experiments on the public datasets IDRiD and DDR, and the experimental results show that the proposed method achieves better performance than other methods.

Texture graph transformer for prostate cancer classification

Prostate cancer classification plays a pivotal role in the diagnosis and treatment of this disease. In this paper, we present a novel approach called the texture graph transformer, which combines texture analysis techniques, graph-based representations, and multi-head attention to enhance the accuracy of prostate cancer classification. Our texture graph transformer adeptly captures intricate texture details and patterns within prostate cancer lesion regions, efficiently improving the model’s ability to discern between different cancer types. By leveraging graph-based representations, our model constructs texture graphs that effectively capture the spatial and textural relationships among various regions in MRI images, enabling comprehensive feature extraction. The incorporation of a texture graph multi-head attention further enhances the model’s capacity to capture long-range dependencies and global context, with a specific focus on informative regions. Through extensive experiments conducted on a dataset comprising 226 MRI images, we demonstrate the efficacy of our proposed approach in significantly improving the performance of prostate cancer classification.

CFNet: Automatic multi-modal brain tumor segmentation through hierarchical coarse-to-fine fusion and feature communication

Automatic segmentation of brain tumors employing images from multi-modalities is important for preoperative diagnosis and prognostic assessment. The rich complementary information contained within multi-modal images allows for improved brain tumor segmentation performance when models are trained on multi-modal data. However, accurate segmentation of small lesion regions from medical images remains challenging, due to the irregular shapes and low boundary contrast of brain gliomas. To address these challenges, we propose a coarse-to-fine feature fusion network (CFNet) that effectively incorporates multi-modal image features through modal interaction, semantic perception, and feature fusion. Specifically, the modality cross-attention fusion module is proposed introducing complementary features and mapping them into a unified semantic space to learn complementary representations between modalities. Furthermore, the multi-scale context perception module enables the model to focus on multi-scale lesion fine-grained information while capturing deep semantics. Moreover, the further fusion of fine-grained features with coarse-grained features enhances edge features and the complementary of features. We evaluate the proposed CFNet on two public brain tumor segmentation datasets BraTS2019 and BraTS2020. Experimental results show that the proposed framework outperforms state-of-the-art methods in Dice score, HD95, Sensitivity, demonstrating the effectiveness of CFNet on brain tumor segmentation. Code will be available at https://github.com/YaruC/CFNet.

Elucidating the hemodynamic impact of residual stenosis post-carotid artery stenting: A numerical study.

Residual stenosis (RS) and hemodynamics demonstrate a significant correlation with postoperative in-stent restenosis/thrombosis following carotid artery stenting (CAS). This study endeavors to elucidate the potential associations between RS and adverse postoperative hemodynamic factors. This study utilized 46 patient-specific carotid artery models post-stenting, which were categorized into two groups based on the presence of RS: the normal group (N, n=23) and the RS group (RS, n=23). A comparative analysis was conducted to evaluate the discrepancies in geometry and adverse hemodynamic parameters, alongside investigating the potential correlation between hemodynamic and geometric parameters. The results reveal that a higher reflux flow volume is discernible in the RS group during low-velocity phases of the cardiac cycle, concomitant with an augmented extent of areas exposed to oscillatory shear stress and extended particle residence time. Moreover, the adverse hemodynamic parameters exhibit a positive correlation with the degree of stent expansion, stent length in the common carotid artery (CCA), and the distal slope of the RS. The distal slope and tortuosity of RS significantly influence the development of adverse hemodynamic conditions post-stenting, exacerbating the hemodynamic environment near the stenosis. Interestingly, while an extended stent length in the internal carotid artery (ICA) region improves hemodynamics by reducing flow disturbance, a longer stent in the CCA significantly worsens these conditions. Hence, it is prudent to analyze the characteristics of the local lesion regions to optimize the strategy for stent implantation.

ECF-Net: Enhanced, Channel-Based, Multi-Scale Feature Fusion Network for COVID-19 Image Segmentation

Accurate segmentation of COVID-19 lesion regions in lung CT images aids physicians in analyzing and diagnosing patients’ conditions. However, the varying morphology and blurred contours of these regions make this task complex and challenging. Existing methods utilizing Transformer architecture lack attention to local features, leading to the loss of detailed information in tiny lesion regions. To address these issues, we propose a multi-scale feature fusion network, ECF-Net, based on channel enhancement. Specifically, we leverage the learning capabilities of both CNN and Transformer architectures to design parallel channel extraction blocks in three different ways, effectively capturing diverse lesion features. Additionally, to minimize irrelevant information in the high-dimensional feature space and focus the network on useful and critical information, we develop adaptive feature generation blocks. Lastly, a bidirectional pyramid-structured feature fusion approach is introduced to integrate features at different levels, enhancing the diversity of feature representations and improving segmentation accuracy for lesions of various scales. The proposed method is tested on four COVID-19 datasets, demonstrating mIoU values of 84.36%, 87.15%, 83.73%, and 75.58%, respectively, outperforming several current state-of-the-art methods and exhibiting excellent segmentation performance. These findings provide robust technical support for medical image segmentation in clinical practice.

Efficient skin lesion segmentation with boundary distillation.

Medical image segmentation models are commonly known for their complex structures, which often render them impractical for use on edge computing devices and compromising efficiency in the segmentation process. In light of this, the industry has proposed the adoption of knowledge distillation techniques. Nevertheless, the vast majority of existing knowledge distillation methods are focused on the classification tasks of skin diseases. Specifically, for the segmentation tasks of dermoscopy lesion images, these knowledge distillation methods fail to fully recognize the importance of features in the boundary regions of lesions within medical images, lacking boundary awareness for skin lesions. This paper introduces pioneering medical image knowledge distillation architecture. The aim of this method is to facilitate the efficient transfer of knowledge from existing complex medical image segmentation networks to a more simplified student network. Initially, a masked boundary feature (MBF) distillation module is designed. By applying random masking to the periphery of skin lesions, the MBF distillation module obliges the student network to reproduce the comprehensive features of the teacher network. This process, in turn, augments the representational capabilities of the student network. Building on the MBF distillation module, this paper employs a cascaded combination approach to integrate the MBF distillation module into a multi-head boundary feature (M2BF) distillation module, further strengthening the student network's feature learning ability and enhancing the overall image segmentation performance of the distillation model. This method has been experimentally validated on the public datasets ISIC-2016 and PH2, with results showing significant performance improvements in the student network. Our findings highlight the practical utility of the lightweight network distilled using our approach, particularly in scenarios demanding high operational speed and minimal storage usage. This research offers promising prospects for practical applications in the realm of medical image segmentation.

Nonrigid registration method for longitudinal chest CT images in COVID-19

Rationale and objectivesTo analyze morphological changes in patients with COVID-19-associated pneumonia over time, a nonrigid registration technique is required that reduces differences in respiratory phase and imaging position and does not excessively deform the lesion region. A nonrigid registration method using deep learning was applied for lung field alignment, and its practicality was verified through quantitative evaluation, such as image similarity of whole lung region and image similarity of lesion region, as well as visual evaluation by a physician. Materials and methodsFirst, the lung field positions and sizes of the first and second CT images were roughly matched using a classical registration method based on iterative calculations as a preprocessing step. Then, voxel-by-voxel transformation was performed using VoxelMorph, a nonrigid deep learning registration method. As an objective evaluation, the similarity of the images was calculated. To evaluate the invariance of image features in the lesion site, primary statistics and 3D shape features were calculated and statistically analyzed. Furthermore, as a subjective evaluation, the similarity of images and whether nonrigid transformation caused unnatural changes in the shape and size of the lesion region were visually evaluated by a pulmonologist. ResultsThe proposed method was applied to 509 patient data points with high image similarity. The variances in histogram characteristics before and after image deformation were confirmed. Visual evaluation confirmed the agreement between the shape and internal structure of the lung field and the natural deformation of the lesion region. ConclusionThe developed nonrigid registration method was shown to be effective for quantitative time series analysis of the lungs.

A high-order focus interaction model and oral ulcer dataset for oral ulcer segmentation

Computer-aided diagnosis has been slow to develop in the field of oral ulcers. One of the major reasons for this is the lack of publicly available datasets. However, oral ulcers have cancerous lesions and their mortality rate is high. The ability to recognize oral ulcers at an early stage in a timely and effective manner is a very critical issue. In recent years, although there exists a small group of researchers working on these, the datasets are private. Therefore to address this challenge, in this paper a multi-tasking oral ulcer dataset (Autooral) containing two major tasks of lesion segmentation and classification is proposed and made publicly available. To the best of our knowledge, we are the first team to make publicly available an oral ulcer dataset with multi-tasking. In addition, we propose a novel modeling framework, HF-UNet, for segmenting oral ulcer lesion regions. Specifically, the proposed high-order focus interaction module (HFblock) performs acquisition of global properties and focus for acquisition of local properties through high-order attention. The proposed lesion localization module (LL-M) employs a novel hybrid sobel filter, which improves the recognition of ulcer edges. Experimental results on the proposed Autooral dataset show that our proposed HF-UNet segmentation of oral ulcers achieves a DSC value of about 0.80 and the inference memory occupies only 2029 MB. The proposed method guarantees a low running load while maintaining a high-performance segmentation capability. The proposed Autooral dataset and code are available from https://github.com/wurenkai/HF-UNet-and-Autooral-dataset.