Dice Similarity Coefficient Research Articles

Longitudinal Magnetic Resonance Imaging in Asymptomatic C9orf72 Mutation Carriers Distinguishes Phenoconverters to Amyotrophic Lateral Sclerosis or Amyotrophic Lateral Sclerosis With Frontotemporal Dementia.

We prospectively studied asymptomatic C9orf72 mutation carriers, identifying those developing amyotrophic lateral sclerosis (ALS) or frontotemporal dementia (FTD). We enrolled 56 asymptomatic family members (AFM) with a C9orf72 mutation (AFM C9+), 132 non-carriers (AFM C9-), and 359 population-based controls. Using 3 T magnetic resonance imaging, we measured cortical thickness, gyrification, and subcortical volumes longitudinally. Linear mixed-effects models on non-converting AFM C9+ scans (n = 107) created a reference for these measurements, establishing individual atrophy patterns. Atrophy patterns from presymptomatic phenoconverters (n = 10 scans) served as a template for group comparisons and similarity assessments. Similarity with phenoconverters was quantified using Dice similarity coefficient (DSC) for cortical and Kullback-Leibler similarity (KLS) for subcortical measures. Using longitudinal similarity assessments, we predicted when participants would reach the average similarity level of phenoconverters at their first post-onset scan. Five AFM C9+ converted to ALS or ALS-FTD. Up to 6 years before symptoms, these phenoconverters exhibited significant atrophy in frontal, temporal, parietal, and cingulate cortex, along with smaller thalamus, hippocampus, and amygdala compared to other AFM C9+. Some non-converted AFM C9+ had high DSC and KLS, approaching values of phenoconverters, whereas others, along with AFM C9- and controls, had lower values. At age 80, we predicted 27.9% (95% confidence interval, 13.2-40.1%) of AFM C9+ and no AFM C9- would reach the same DSC as phenoconverters. Distinctive atrophy patterns are visible years before symptom onset on presymptomatic scans of phenoconverters. Combining baseline and follow-up similarity measures may serve as a promising imaging biomarker for identifying those at risk of ALS or ALS-FTD. ANN NEUROL 2024.

Curved multiplanar reformatting allows the accurate histological delineation of hippocampal subfields.

Hippocampal subfields perform specific roles in normal cognitive functioning and have distinct vulnerabilities in neurological disorders. However, measurement of subfields with MRI is technically difficult in the head and tail of the hippocampus. Recent studies have utilized curved multiplanar reconstruction (CMPR) to improve subfield visualization in the head and tail, but this method has not yet been applied to histological data. We utilized BigBrain data, an open-source database of serially sectioned histological data for our analyses. The left hippocampus was segmented according to histological criteria by two raters in order to evaluate intra- and inter-rater reliability of histology-based segmentation throughout the long axis. Segmentation according to our previous protocol for the hippocampal body was then compared to these histological measurements to evaluate for histological validity. Agreement between segmentations was evaluated using Dice similarity coefficients (DSCs). Intra-rater reliability (DSCs) of histological segmentation was excellent for all subfields: CA1 (0.8599), CA2 (0.7586), CA3/CA4/DG (0.8907), SLM (0.9123), subiculum (0.8149). Similarly, inter-rater reliability analysis demonstrated excellent agreement (DSCs) for all subfield locations: CA1 (0.8203), CA2 (0.7253), CA3/CA4/DG (0.8439), SLM (0.8700), subiculum (0.7794). Finally, histological accuracy (DSCs) for our previous protocol was excellent for all subfields: CA1 (0.8821), CA2 (0.8810), CA3/CA4/DG (0.9802), SLM (0.9879), subiculum (0.8774). When subfields in the hippocampus head, body, and tail were analyzed independently, DSCs also showed excellent agreement. CMPR allows reliable subfield segmentation based on histological criteria throughout the hippocampal head, body, and tail. Our previous protocol for the hippocampal body can be applied to provide histologically valid subfield measurements throughout the entire hippocampal long axis.

Impact of Artificial Intelligence-Based Autosegmentation of Organs at Risk in Low- and Middle-Income Countries

Radiation therapy (RT) processes require significant human resources and expertise, creating a barrier to rapid RT deployment in low- and middle-income countries (LMICs). Accurate segmentation of tumor targets and organs at risk (OARs) is crucial for optimal RT. This study assessed the impact of artificial intelligence (AI)-based autosegmentation of OARs in 2 LMICs. Ten patients, comprising 5 head and neck (HN) cancer patients and 5 prostate cancer patients, were randomly selected. Planning computed tomography images were subjected to autosegmentation using an Food and Drug Administration-approved AI software tool and manual segmentation by experienced radiation oncologists from 2 LMIC RT clinics. The control data, obtained from a large academic institution in the United States, consisted of contours obtained by an experienced radiation oncologist. The segmentation time, DICE similarity coefficient (DSC), Hausdorff distance, and mean surface distance were evaluated. AI significantly reduced segmentation time, averaging 2 minutes per patient, compared with 57 to 84 minutes for manual contouring in LMICs. Compared with the control data, the AI pelvic contours provided better agreement than did the LMIC manual contours (mean DSC of 0.834 vs 0.807 in LMIC1 and 0.844 vs 0.801 in LMIC2). For HN contours, AI provided better agreement for the majority of OAR contours than manual contours in LMIC1 (mean DSC: 0.823 vs 0.821) or LMIC2 (mean DSC: 0.792 vs 0.748). Neither the AI nor LMIC manual contours had good agreement with the control data (DSC < 0.600) for the optic nerves, chiasm, and cochlea. AI-based autosegmentation generates OAR contours of comparable quality to manual segmentation for both pelvic and HN cancer patients in LMICs, with substantial time savings.

Deformable Mapping of Rectal Cancer Whole-Mount Histology with Restaging MRI at Voxel Scale: A Feasibility Study.

Purpose To develop a radiology-pathology coregistration method for 1:1 automated spatial mapping between preoperative rectal MRI and ex vivo rectal whole-mount histology (WMH). Materials and Methods This retrospective study included consecutive patients with rectal adenocarcinoma who underwent total neoadjuvant therapy followed by total mesorectal excision with preoperative rectal MRI and WMH from January 2019 to January 2022. A gastrointestinal pathologist and a radiologist established three corresponding levels for each patient at rectal MRI and WMH, subsequently delineating external and internal rectal wall contours and the tumor bed at each level and defining eight point-based landmarks. An advanced deformable image coregistration model based on the linearized iterative boundary reconstruction (LIBR) approach was compared with rigid point-based registration (PBR) and state-of-the-art deformable intensity-based multiscale spectral embedding registration (MSERg). Dice similarity coefficient (DSC), modified Hausdorff distance (MHD), and target registration error (TRE) across patients were calculated to assess the coregistration accuracy of each method. Results Eighteen patients (mean age, 54 years ± 13 [SD]; nine female) were included. LIBR demonstrated higher DSC versus PBR for external and internal rectal wall contours and tumor bed (external: 0.95 ± 0.03 vs 0.86 ± 0.04, respectively, P < .001; internal: 0.71 ± 0.21 vs 0.61 ± 0.21, P < .001; tumor bed: 0.61 ± 0.17 vs 0.52 ± 0.17, P = .001) and versus MSERg for internal rectal wall contours (0.71 ± 0.21 vs 0.63 ± 0.18, respectively; P < .001). LIBR demonstrated lower MHD versus PBR for external and internal rectal wall contours and tumor bed (external: 0.56 ± 0.25 vs 1.68 ± 0.56, respectively, P < .001; internal: 1.00 ± 0.35 vs 1.62 ± 0.59, P < .001; tumor bed: 2.45 ± 0.99 vs 2.69 ± 1.05, P = .03) and versus MSERg for internal rectal wall contours (1.00 ± 0.35 vs 1.62 ± 0.59, respectively; P < .001). LIBR demonstrated lower TRE (1.54 ± 0.39) versus PBR (2.35 ± 1.19, P = .003) and MSERg (2.36 ± 1.43, P = .03). Computation time per WMH slice for LIBR was 35.1 seconds ± 12.1. Conclusion This study demonstrates feasibility of accurate MRI-WMH coregistration using the advanced LIBR method. Keywords: MR Imaging, Abdomen/GI, Rectum, Oncology Supplemental material is available for this article. © RSNA, 2024.

Use of Ultra-Short Echo Time MRI to Improve Temporal Bone Imaging.

The short T2 nature of cortical bone causes it to appear similar to air on MR, forcing clinicians to rely on computed tomography imaging, with its attendant ionizing radiation exposure, to define temporal bone structures. Through the use of novel MR sequences with ultra-short echo times (UTE), short T2 structures are now able to be visualized, allowing for improved understanding of anatomical relationships. Eight patients (50% female) undergoing MR imaging of the skull base for diagnostic purposes (62.5% for vestibular schwannoma surveillance) at a tertiary care center were enrolled to evaluate the safety and efficacy of UTE imaging. CT scans were completed in 37.5% of the patients as part of their workup and used for comparison purposes. The repetition time, short echo time, and long echo time for the UTE sequence were 11, 0.032, and 2.2 msec, respectively. The protocol added 6 min to the total scanning time, and all patients tolerated the sequence without issue. The ossicles, mastoid air cells, antrum, and epitympanum were able to be seen and had a high Dice similarity coefficient when compared to CT (>0.5). UTE allowed for clear delineation of all segments of the facial nerve with a signal-to-noise ratio of 35 (although the BRAVO sequences had a superior ratio of 140). Vestibular schwannomas were able to be distinguished from normal brain parenchyma. UTE is safe and effective for visualizing anatomic structures not normally seen on traditional MRI, potentially allowing for improved surgical planning in patients. 3 Laryngoscope, 134:4691-4696, 2024.

Echocardiographic mitral valve segmentation model

Segmentation of mitral valve is not only important for clinical diagnosis, but also has far-reaching impact on prevention and prognosis of the disease by experts and doctors.In this paper, the multi-channel cross fusion transformer based U-Net network model (MCCT-UNet) is proposed according to the classical U-Net architecture. First, the jump connection part of MCCT-UNet is designed by using a multi-channel cross-fusion based attention mechanism module (MCCT) instead of the original jump connection, and this module fuses the feature maps from different scales in different stages of the encoder. Second, the optimization of the feature fusion method is proposed in the decoding stage by designing the cross-compression excitation sub-module (C-SENet) to replace the simple feature splicing, and the C-SENet is used to bridge the inconsistency of the semantic hierarchy by effectively combining the deeper information in the encoding stage with the shallower information. This two modules can establish a close connection between the encoder and decoder by exploring multi-scale global contextual information to solve the semantic divide problem, thus it significantly improves the segmentation performance of the network. The experimental results show that the improvement is effective, and the MCCT-UNet model outperforms the other 9 network models. Specifically, the MCCT-UNet achieved a Dice coefficient of 0.8734, an IoU of 0.7854, and an accuracy of 0.9977, demonstrating significant improvements over the compared models.

A flexible 2.5D medical image segmentation approach with in-slice and cross-slice attention

Deep learning has become the de facto method for medical image segmentation, with 3D segmentation models excelling in capturing complex 3D structures and 2D models offering high computational efficiency. However, segmenting 2.5D images, characterized by high in-plane resolution but lower through-plane resolution, presents significant challenges. While applying 2D models to individual slices of a 2.5D image is feasible, it fails to capture the spatial relationships between slices. On the other hand, 3D models face challenges such as resolution inconsistencies in 2.5D images, along with computational complexity and susceptibility to overfitting when trained with limited data. In this context, 2.5D models, which capture inter-slice correlations using only 2D neural networks, emerge as a promising solution due to their reduced computational demand and simplicity in implementation. In this paper, we introduce CSA-Net, a flexible 2.5D segmentation model capable of processing 2.5D images with an arbitrary number of slices. CSA-Net features an innovative Cross-Slice Attention (CSA) module that effectively captures 3D spatial information by learning long-range dependencies between the center slice (for segmentation) and its neighboring slices. Moreover, CSA-Net utilizes the self-attention mechanism to learn correlations among pixels within the center slice. We evaluated CSA-Net on three 2.5D segmentation tasks: (1) multi-class brain MR image segmentation, (2) binary prostate MR image segmentation, and (3) multi-class prostate MR image segmentation. CSA-Net outperformed leading 2D, 2.5D, and 3D segmentation methods across all three tasks, achieving average Dice coefficients and HD95 values of 0.897 and 1.40 mm for the brain dataset, 0.921 and 1.06 mm for the prostate dataset, and 0.659 and 2.70 mm for the ProstateX dataset, demonstrating its efficacy and superiority. Our code is publicly available at: https://github.com/mirthAI/CSA-Net.

Automatic segmentation and diameter measurement of deep medullary veins.

As one of the pathogenic factors of cerebral small vessel disease, venous collagenosis may result in the occlusion or stenosis of deep medullary veins (DMVs). Although numerous DMVs can be observed in susceptibility-weighted MRI images, their diameters are usually smaller than the MRI resolution, making it difficult to segment them and quantify their sizes. We aim to automatically segment DMVs and measure their diameters from gradient-echo images. A neural network model was trained for DMV segmentation based on the gradient-echo magnitude and phase images of 20 subjects at 7 T. The diameters of DMVs were obtained by fitting measured complex images with model images that accounted for the DMV-induced magnetic field and point spread function. A phantom study with graphite rods of different diameters was conducted to validate the proposed method. Simulation was carried out to evaluate the voxel-size dependence of measurement accuracy for a typical DMV size. The automatically segmented DMV masks had Dice similarity coefficients of 0.68 ± 0.03 (voxel level) and 0.83 ± 0.04 (cluster level). The fitted graphite-rod diameters closely matched their true values. In simulation, the fitted diameters closely matched the true value when voxel size was ≤ 0.45 mm, and 92.2% of DMVs had diameters between 90 μm and 200 μm with a peak at about 120 μm, which agreed well with an earlier ex vivo report. The proposed methods enabled efficient and quantitative study of DMVs, which may help illuminate the role of DMVs in the etiopathogenesis of cerebral small vessel disease.

Prediction of femoral head collapse in osteonecrosis using deep learning segmentation and radiomics texture analysis of MRI

BackgroundFemoral head collapse is a critical pathological change and is regarded as turning point in disease progression in osteonecrosis of the femoral head (ONFH). In this study, we aim to build an automatic femoral head collapse prediction pipeline for ONFH based on magnetic resonance imaging (MRI) radiomics.MethodsIn the segmentation model development dataset, T1-weighted MRI of 222 hips from two hospitals were retrospectively collected and randomly split into training (n = 190) and test (n = 32) sets. In the prognosis prediction model development dataset, 206 hips were also retrospectively collected from two hospitals and divided into training set (n = 155) and external test set (n = 51) according to data source. A deep learning model for automatic lesion segmentation was trained with nnU-Net, from which three-dimensional regions of interest were segmented and a total of 107 radiomics features were extracted. After intra-class correlation coefficients screening, feature correlation coefficient screening and Least Absolute Shrinkage and Selection Operator regression feature selection, a machine learning model for ONFH prognosis prediction was trained with Logistic Regression (LR) and Light Gradient Boosting Machine (LightGBM) algorithm.ResultsThe segmentation model achieved an average dice similarity coefficient of 0.848 and an average 95% Hausdorff distance of 3.794 in the test set, compared to the manual segmentation results. After feature selection, nine radiomics features were included in the prognosis prediction model. External test showed that the LightGBM model exhibited acceptable predictive performance. The area under the curve (AUC) of the prediction model was 0.851 (95% CI: 0.7268–0.9752), with an accuracy of 0.765, sensitivity of 0.833, and specificity of 0.727. Decision curve analysis showed that the LightGBM model exhibited favorable clinical utility.ConclusionThis study presents an automated pipeline for predicting femoral head collapse in ONFH with acceptable performance. Further research is necessary to determine the clinical applicability of this radiomics-based approach and to assess its potential to assist in treatment decision-making for ONFH.

Effective encoder-decoder network for pupil light reflex segmentation in facial photographs of ptosis patients

Accurate segmentation of pupil light reflexes is essential for the reliable assessment of ptosis severity, a condition characterized by the drooping of the upper eyelid. This study introduces a novel encoder-decoder network specialized in reflex segmentation by focusing on addressing issues related to very small regions of interest from an architectural perspective. Specifically, the proposed network is designed to exploit low-level features effectively by integrating a multi-level skip connection and a 1 × 1 convolution-enhanced initial encoding stage. Assessed using a photograph image dataset from Chung-Ang University Hospital, which includes 87 healthy subjects, 64 with ptosis, and 257 with Graves’ orbitopathy (collected between January 2010 and February 2023), the proposed network outperforms five conventional encoder-decoders. Over 30 trials, the proposed network achieved a mean Dice coefficient of 0.767 and an Intersection over Union of 0.653, indicating a statistically significant improvement in the segmentation of reflex. Our findings show that an elaborate design based on the lowest-level skip connection and 1 × 1 convolution at initial stage enhances the segmentation of pupil light reflexes. The source code of the proposed network is available at https://github.com/tkdgur658/ReflexNet.

Automated Organ Segmentation for Radiation Therapy: A Comparative Analysis of AI-Based Tools Versus Manual Contouring in Korean Cancer Patients

Background: Accurate delineation of tumors and organs at risk (OARs) is crucial for intensity-modulated radiation therapy. This study aimed to evaluate the performance of OncoStudio, an AI-based auto-segmentation tool developed for Korean patients, compared with Protégé AI, a globally developed tool that uses data from Korean cancer patients. Methods: A retrospective analysis of 1200 Korean cancer patients treated with radiotherapy was conducted. Auto-contours generated via OncoStudio and Protégé AI were compared with manual contours across the head and neck and thoracic, abdominal, and pelvic organs. Accuracy was assessed using the Dice similarity coefficient (DSC), mean surface distance (MSD), and 95% Hausdorff distance (HD). Feedback was obtained from 10 participants, including radiation oncologists, residents, and radiation therapists, via an online survey with a Turing test component. Results: OncoStudio outperformed Protégé AI in 85% of the evaluated OARs (p &lt; 0.001). For head and neck organs, OncoStudio achieved a similar DSC (0.70 vs. 0.70, p = 0.637) but significantly lower MSD and 95% HD values (p &lt; 0.001). In thoracic organs, OncoStudio performed excellently in 90% of cases, with a significantly greater DSC (male: 0.87 vs. 0.82, p &lt; 0.001; female: 0.95 vs. 0.87, p &lt; 0.001). OncoStudio also demonstrated superior accuracy in abdominal (DSC 0.88 vs. 0.81, p &lt; 0.001) and pelvic organs (male: DSC 0.95 vs. 0.85, p &lt; 0.001; female: DSC 0.82 vs. 0.73, p &lt; 0.001). Clinicians favored OncoStudio in 70% of cases, with 90% endorsing its clinical suitability for Korean patients. Conclusions: OncoStudio, which is tailored for Korean patients, demonstrated superior segmentation accuracy across multiple anatomical regions, suggesting its suitability for radiotherapy planning in this population.

Automatic tumor segmentation and lymph node metastasis prediction in papillary thyroid carcinoma using ultrasound keyframes.

Accurate preoperative prediction of cervical lymph node metastasis (LNM) for papillary thyroid carcinoma (PTC) patients is essential for disease staging and individualized treatment planning, which can improve prognosis and facilitate better management. To establish a fully automated deep learning-enabled model (FADLM) for automated tumor segmentation and cervical LNM prediction in PTC using ultrasound (US) video keyframes. The bicentral study retrospective enrolled 518 PTC patients, who were then randomly divided into the training (Hospital 1, n=340), internal test (Hospital 1, n=83), and external test cohorts (Hospital 2, n=95). The FADLM integrated mask region-based convolutional neural network (Mask R-CNN) for automatic thyroid primary tumor segmentation and ResNet34 with Bayes strategy for cervical LNM diagnosis. A radiomics model (RM) using the same automated segmentation method, a traditional radiomics model (TRM) using manual segmentation, and a clinical-semantic model (CSM) were developed for comparison. The dice similarity coefficient (DSC) was used to evaluate segmentation performance. The prediction performance of the models was validated in terms of discrimination and clinical utility with the area under the receiver operator characteristic curve (AUC), heatmap analysis, and decision curve analysis (DCA). The comparison of the predictive performance among different models was conducted by DeLong test. The performances of two radiologists compared with FADLM and the diagnostic augmentation with FADLM's assistance were analyzed in terms of accuracy, sensitivity and specificity using McNemar's x2 test. The p-value less than 0.05 was defined as a statistically significant difference. The Benjamini-Hochberg procedure was applied for multiple comparisons to deal with Type I error. The FADLM yielded promising segmentation results in training (DSC: 0.88±0.23), internal test (DSC: 0.88±0.23), and external test cohorts (DSC: 0.85±0.24). The AUCs of FADLM for cervical LNM prediction were 0.78 (95% CI: 0.73, 0.83), 0.83 (95% CI: 0.74, 0.92), and 0.83 (95% CI: 0.75, 0.92), respectively. It all significantly outperformed the RM (AUCs: 0.78vs. 0.72; 0.83vs. 0.65; 0.83vs. 0.68, all adjusted p-values<0.05) and CSM (AUCs: 0.78vs. 0.71; 0.83vs. 0.62; 0.83vs. 0.68, all adjusted p-values<0.05) across the three cohorts. The RM offered similar performance to that of TRM (AUCs: 0.61vs. 0.63, adjusted p-value =0.60) while significantly reducing the segmentation time (3.3±3.8 vs. 14.1±4.2 s, p-value <0.001). Under the assistance of FADLM, the accuracies of junior and senior radiologists were improved by 18% and 15% (all adjusted p-values<0.05) and the sensitivities by 25% and 21% (all adjusted p-values<0.05) in the external test cohort. The FADLM with elaborately designed automated strategy using US video keyframes holds good potential to provide an efficient and consistent prediction of cervical LNM in PTC. The FADLM displays superior performance to RM, CSM, and radiologists with promising efficacy.

Development and Validation of Deep Learning Preoperative Planning Software for Automatic Lumbosacral Screw Selection Using Computed Tomography

Achieving precise pedicle screw placement in posterior lumbar interbody fusion (PLIF) is essential but difficult due to the intricacies of manual preoperative planning with CT scans. We analyzed CT data from 316 PLIF patients, using Mimics software for manual planning by two surgeons. A deep learning model was trained on 228 patients and validated on 88 patients, assessing planning efficiency and accuracy. Automatic planning successfully segmented and placed screws in all 316 cases, significantly outperforming manual planning in speed. The Dice coefficient for segmentation accuracy was 0.95. The difference in mean pedicle transverse angle (PTA) and pedicle sagittal angle (PSA) for automatic planning screws compared to manual planning screws was 1.63 ± 0.83° and 1.39 ± 1.03°, respectively, and these differences were either statistically comparable or not significantly different compared to the variability of manual planning screws. The average Dice coefficient of implanted screws was 0.63 ± 0.08, and the consistency between automatic screws and manual reference screws was higher than that of internal screws (Dice 0.62 ± 0.09). Compared with manual screws, automatic screws were shorter (46.58 ± 3.09 mm) and thinner (6.24 ± 0.35 mm), and the difference was statistically significant. In qualitative validation, 97.7% of the automatic planning screws were rated Gertzbein–Robbins (GR) Class A and 97.3% of the automatic planning screws were rated Badu Class 0. Deep learning software automates lumbosacral pedicle screw planning, enhancing surgical efficiency and accuracy.

Deep learning-based segmentation of kidneys and renal cysts on T2-weighted MRI from patients with autosomal dominant polycystic kidney disease

BackgroundOur aim was to train and test a deep learning-based algorithm for automatically segmenting kidneys and renal cysts in patients with autosomal dominant polycystic kidney disease (ADPKD).MethodsWe retrospectively selected all ADPKD patients who underwent renal MRI with coronal T2-weighted imaging at our institution from 2008 to 2022. The 20 most recent examinations constituted the test dataset, to mimic pseudoprospective enrolment. The remaining ones constituted the training dataset to which eight normal renal MRIs were added. Kidneys and cysts ground truth segmentations were performed on coronal T2-weighted images by a junior radiologist supervised by an experienced radiologist. Kidneys and cysts of the 20 test MRIs were segmented by the algorithm and three independent human raters. Segmentations were compared using overlap metrics. The total kidney volume (TKV), total cystic volume (TCV), and cystic index (TCV divided by TKV) were compared using Bland–Altman analysis.ResultsWe included 164 ADPKD patients. Dice similarity coefficients ranged from 85.9% to 87.4% between the algorithms and the raters’ segmentations and from 84.2% to 86.2% across raters’ segmentations. For TCV assessment, the biases ± standard deviations (SD) were 3–19 ± 137–151 mL between the algorithm and the raters, and 22–45 ± 49–57 mL across raters. The algorithm underestimated TKV and TCV in two outliers with TCV > 2800 mL. For cystic index assessment, the biases ± SD were 2.5–6.9% ± 6.7–8.3% between the algorithm and the raters, and 2.1–9.4 ± 7.4–11.6% across raters.ConclusionThe algorithm’s performance fell within the range of inter-rater variability, but large TKV and TCV were underestimated.Relevance statementAccurate automated segmentation of the renal cysts will enable the large-scale evaluation of the prognostic value of TCV and cystic index in ADPKD patients. If these biomarkers are prognostic, then automated segmentation will facilitate their use in daily routine.Key PointsCystic volume is an emerging biomarker in ADPKD.The algorithm’s performance in segmenting kidneys and cysts fell within interrater variability.The segmentation of very large cysts, under-represented in the training dataset, needs improvement.Graphical

PZS‐Net: Incorporating of Frame Sequence and Multi‐Scale Priors for Prostate Zonal Segmentation in Transrectal Ultrasound

Transrectal ultrasound (TRUS) videos offer valuable histopathologic information about the prostate. Accurate prostate zonal segmentation in TRUS videos is vital for diagnosing prostate cancer and guiding surgery. However, TRUS videos are manually recorded by urologists, resulting in no standardized coordinate system, which limits direct prostate zonal segmentation in these videos. To overcome the limitation, a novel Prostate Zonal Segmentation Network (PZS‐Net), based on U‐Net, which learns critical cross‐frame information and multi‐scale features from sequential frames, is proposed. First, a sequential frame cross‐attention (SFCA) module is designed to capture remote information from sequential frames to enhance the feature representation of the current frame. The SFCA module is embedded at each skip connection layer to extract crucial cross‐frame information. Then, a multi‐scale fusion (MSF) module that utilizes three parallel branches with different atrous convolutions is designed. The MSF module is placed at the bottleneck layer to dynamically fuse multi‐scale context information from high‐level features. Extensive experiments on TRUS image datasets show that the PZS‐Net achieves higher accuracy in both the transitional zone (dice coefficient [Dice]: 68.90% ± 1.73%, mean intersection over union [mIoU]: 59.19% ± 2.09%, 95% Hausdorff distance [HD95]: 5.02 ± 0.83 mm) and the peripheral zone (Dice: 63.99% ± 3.16%, mIoU: 54.60% ± 3.35%, HD95: 5.28 ± 1.12 mm) and demonstrates the effectiveness and competitiveness of its key components via comprehensive ablation studies.

MRS‐Net: Brain tumour segmentation network based on feature fusion and attention mechanism

AbstractAccurate segmentation of brain tumor magnetic resonance imaging (MRI) is crucial for treatment planning. Addressing the challenges of complex tumor structures and inadequate cross‐channel information utilization in Unet‐based segmentation, this paper proposes the multi‐scale residual brain tumor MRI segmentation network (MRS‐Net) incorporating an attention mechanism to enhance segmentation accuracy. First, the double residual feature fusion module is utilized to enhance the fusion of feature information between different levels. Second, the Atrous Spatial Pyramid Pooling is introduced as a bridging module of the network to capture the features at different scales of the image, so as to enhance the extraction capability of the network for detailed features. Finally, the inverted residual coordinate attention module replaces the direct splicing in Unet to fuse the large feature information at each level and scale, thus enhancing the model's ability to recognize the spatial location information of brain tumors. The Dice coefficients, positive predictive values (PPVs), sensitivities (Sensitivity) and Hausdorff distance (HD), which are the four evaluation indexes, reach 84.54%, 87.43%, 88.37% and 2.248, respectively, which are improved by 1.85%, 2.11%, 2.88% and 6.0%, respectively, compared with Unet. The experimental results show that MRS‐Net achieves better brain tumor image segmentation.

Deep learning for quantification of myocardial scar and microvascular obstruction in late gadolinium enhancement cardiovascular magnetic resonance images

Abstract Objectives Detecting and measuring myocardial scar and microvascular obstruction (MVO) accurately following reperfusion therapy is clinically vital for patients with acute myocardial infarction (AMI). Current clinical practice relies on manual delineation of endocardial borders as well as hyper- and hypo-enhanced regions on late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR) images for scar and MVO quantification, respectively. However, this approach is subjective, labor-intensive, and time-consuming. The objective of this study is to develop and assess a convolutional neural network (CNN)-based method for the automated quantification of LGE scar and MVO in AMI patients. Methods A total of 1836 LGE images from 252 patients with AMI were analyzed, with the cohort comprising 232 males and an average age of 56±9 years. Our approach, utilizing a combined architecture of cascaded YOLOv8 (You Only Look Once version 8) and nnU-Net, was implemented for the simultaneous segmentation of the left ventricular (LV) myocardium, scar, and MVO in LGE images. Independent datasets for training, validation, and testing were maintained in an 8:1:1 ratio. Performance evaluation was conducted by comparing the results to manual scar and MVO quantification, utilizing metrics such as the Dice similarity coefficient (DSC) and Bland-Altman analysis. Results The automated segmentation process took 0.05 seconds per image. The segmentation accuracy, as assessed by the DSC, yielded values of 0.96±0.04 (per-patient, n=25) and 0.96±0.05 (per-slice, n=184) for scar segmentation, and 0.86±0.17 (per-patient) and 0.89±0.15 (per-slice) for MVO segmentation. In per-patient analysis, the volumes of automatically segmented scar and MVO demonstrated strong agreement with manually derived results, with biases [limits of agreement] of -0.3 [-3.0, 2.4] cm³ and 0.1 [-0.2, 0.3] cm³, respectively. Furthermore, there was strong concordance between automatically and manually derived scar size as a percentage of LV mass (-0.2 [-3.1, 2.7] %) and MVO size as a percentage of scar (1.2 [-5.6, 8.0] %). Conclusions The findings demonstrate excellent agreement between our proposed automated deep-learning solution and clinical expert annotation, suggesting its potential as a predictive tool for prognosis and clinical outcomes in patients with AMI.

Quantitative texture analysis using machine learning for predicting interpretable pulmonary perfusion from non-contrast computed tomography in pulmonary embolism patients

BackgroundPulmonary embolism (PE) is life-threatening and requires timely and accurate diagnosis, yet current imaging methods, like computed tomography pulmonary angiography, present limitations, particularly for patients with contraindications to iodinated contrast agents. We aimed to develop a quantitative texture analysis pipeline using machine learning (ML) based on non-contrast thoracic computed tomography (CT) scans to discover intensity and textural features correlated with regional lung perfusion (Q) physiology and pathology and synthesize voxel-wise Q surrogates to assist in PE diagnosis.MethodsWe retrospectively collected 99mTc-labeled macroaggregated albumin Q-SPECT/CT scans from patients suspected of PE, including an internal dataset of 76 patients (64 for training, 12 for testing) and an external testing dataset of 49 patients. Quantitative CT features were extracted from segmented lung subregions and underwent a two-stage feature selection pipeline. The prior-knowledge-driven preselection stage screened for robust and non-redundant perfusion-correlated features, while the data-driven selection stage further filtered features by fitting ML models for classification. The final classification model, trained with the highest-performing PE-associated feature combination, was evaluated in the testing cohorts based on the Area Under the Curve (AUC) for subregion-level predictability. The voxel-wise Q surrogate was then synthesized using the final selected feature maps (FMs) and model score maps (MSMs) to investigate spatial distributions. The Spearman correlation coefficient (SCC) and Dice similarity coefficient (DSC) were used to assess the spatial consistency between FMs or MSMs and Q-SPECT scans.ResultsThe optimal model performance achieved an AUC of 0.863 during internal testing and 0.828 on the external testing cohort. The model identified a combination containing 14 intensity and textural features that were non-redundant, robust, and capable of distinguishing between high- and low-functional lung regions. Spatial consistency assessment in the internal testing cohort showed moderate-to-high agreement between MSMs and reference Q-SPECT scans, with median SCC of 0.66, median DSCs of 0.86 and 0.64 for high- and low-functional regions, respectively.ConclusionsThis study validated the feasibility of using quantitative texture analysis and a data-driven ML pipeline to generate voxel-wise lung perfusion surrogates, providing a radiation-free, widely accessible alternative to functional lung imaging in managing pulmonary vascular diseases.Clinical trial numberNot applicable.

Temporomandibular joint CBCT image segmentation via multi-view ensemble learning network.

Accurate segmentation of the temporomandibular joint (TMJ) from cone beam CT (CBCT) images holds significant clinical value for diagnosing temporomandibular joint osteoarthrosis (TMJOA) and related conditions. Convolutional neural network-based medical image segmentation methods have achieved state-of-the-art performance in various segmentation tasks. However, 3D medical images segmentation requires substantial global context and rich spatial semantic information, demanding much more GPU memory and computational resources. To address these challenges in 3D medical image segmentation, we propose a novel network- the MVEL-Net (Multi-view Ensemble Learning Network) for TMJ CBCT image segmentation. By resampling images along three dimensions, we generate multiple weak learners with different spatial semantic information. A subsequent strong learning network effectively integrates the outputs from these weak learners to achieve more accurate segmentation results. We evaluated our network model using a clinical dataset comprising 88 subjects with TMJ CBCT images. The average Dice similarity coefficient (DSC) was 0.9817 ± 0.0049, the average surface distance was 0.0540 ± 0.0179mm, and the 95% Hausdorff distance was 0.1743 ± 0.0550mm. Our proposed MVEL-Net demonstrates excellent segmentation performance on TMJ from CBCT images, while using fewer GPU memory resources compared to other 3D networks. The effectiveness of this method in capturing spatial context could be leveraged for tasks like organ segmentation from volumetric scans. This may facilitate wider adoption of AI-based solutions for automated analysis of 3D medical images.

Dual-attention transformer-based hybrid network for multi-modal medical image segmentation

Accurate medical image segmentation plays a vital role in clinical practice. Convolutional Neural Network and Transformer are mainstream architectures for this task. However, convolutional neural network lacks the ability of modeling global dependency while Transformer cannot extract local details. In this paper, we propose DATTNet, DualATTentionNetwork, an encoder-decoder deep learning model for medical image segmentation. DATTNet is exploited in hierarchical fashion with two novel components: (1) Dual Attention module is designed to model global dependency in spatial and channel dimensions. (2) Context Fusion Bridge is presented to remix the feature maps with multiple scales and construct their correlations. The experiments on ACDC, Synapse and Kvasir-SEG datasets are conducted to evaluate the performance of DATTNet. Our proposed model shows superior performance, effectiveness and robustness compared to SOTA methods, with mean Dice Similarity Coefficient scores of 92.2%, 84.5% and 89.1% on cardiac, abdominal organs and gastrointestinal poly segmentation tasks. The quantitative and qualitative results demonstrate that our proposed DATTNet attains favorable capability across different modalities (MRI, CT, and endoscopy) and can be generalized to various tasks. Therefore, it is envisaged as being potential for practicable clinical applications. The code has been released on https://github.com/MhZhang123/DATTNet/tree/main.