Adjacent Slices Research Articles

Inferring allele-specific copy number aberrations and tumor phylogeography from spatially resolved transcriptomics.

Analyzing somatic evolution within a tumor over time and across space is a key challenge in cancer research. Spatially resolved transcriptomics (SRT) measures gene expression at thousands of spatial locations in a tumor, but does not directly reveal genomic aberrations. We introduce CalicoST, an algorithm to simultaneously infer allele-specific copy number aberrations (CNAs) and reconstruct spatial tumor evolution, or phylogeography, from SRT data. CalicoST identifies important classes of CNAs-including copy-neutral loss of heterozygosity and mirrored subclonal CNAs-that are invisible to total copy number analysis. Using nine patients' data from the Human Tumor Atlas Network, CalicoST achieves an average accuracy of 86%, approximately 21% higher than existing methods. CalicoST reconstructs a tumor phylogeography in three-dimensional space for two patients with multiple adjacent slices. CalicoST analysis of multiple SRT slices from a cancerous prostate organ reveals mirrored subclonal CNAs on the two sides of the prostate, forming a bifurcating phylogeography in both genetic and physical space.

Implications of calcified nodule and adjacent calcification severity for target lesion revascularization after percutaneous coronary intervention: multi-center intravascular ultrasound imaging study

Abstract Background Calcified nodule (CN) is a distinct plaque phenotype, characterized as protruding nodular calcification penetrating lumen surface with the attached thrombus. Recent studies revealed that a risk of repeat revascularization after PCI was substantially elevated in patients with coronary artery disease (CAD) attributable to CN. Furthermore, pathohistological studies have elucidated mechanistic insights into the formulation and progression of CN, suggesting that it is formulated by the fracture of sheet calcification at its adjacent segment. Purpose To elucidate the association of calcified plaque volume at CN and adjacent lesions with TLR risks in CAD patients receiving PCI by using IVUS imaging. Methods This multicenter study analyzed IVUS imaging at culprit segment was conducted in 204 CAD patients attributable to de novo CN receiving PCI. The calc grade was assigned at each 1-mm cross-sectional frame of CN and culprit segment, respectively (no calc=0, 1-89°=1, 90-179°=2, 180-269°=3, 270-360°=4). Calcium Volume Index (CVI) at both CN and adjacent zone were calculated (Figure; Calcification volume index at CN zone = the sum of calcium scores at CN・averaged lesion slices of CN / CN lesion slices, Calcification volume index at adjacent zone = the sum of calcium scores at adjacent zone・averaged lesion slices of adjacent zone / adjacent zone slices). Predictors of TLR were analyzed by multivariate Cox hazard model. Results CN was successfully treated by DES (82%) and DCB (18%). 34% and 8.3% of subjects required rotablator and orbital atherectomy, respectively. Consequently, final lumen area after PCI was 7.2 ± 0.4 mm2. During the 2.8-year observational period (interquartile range: 2.4 to 3.2 years), TLR was required in 63 patients (30.9%) of study population. CVI at CN and adjacent zone were 9.9 ± 0.4 and 7.0 ± 0.7, respectively. On multivariate analysis, hemodialysis, final lumen area, CVI at both CN and adjacent zone predicted TLR (Figure). Recurrent TLR (two or more TLRs) was found in 23.3% patients who suffered from TLR. Of note, even among patients who required TLR, CVIs at both zones were significantly higher in patients with recurrent TLR than those without (Figure). Conclusion CAD subjects caused by CN exhibited a considerably high-risk TLRs. Our findings underscore that calcification severity at both CN and adjacent segment predicted TLR and is associated with repeated TLR.methodsResults

An Efficient and Robust 3D Medical Image Classification Approach Based on 3D CNN, Time‐Distributed 2D CNN‐BLSTM Models, and mRMR Feature Selection

ABSTRACTThe advent of 3D medical imaging has been a turning point in the diagnosis of various diseases, as voxel information from adjacent slices helps radiologists better understand complex anatomical relationships. However, the interpretation of medical images by radiologists with different levels of expertise can vary and is also time‐consuming. In the last decades, artificial intelligence‐based computer‐aided systems have provided fast and more reliable diagnostic insights with great potential for various clinical purposes. This paper proposes a significant deep learning based 3D medical image diagnosis method. The method classifies MedMNIST3D, which consists of six 3D biomedical datasets obtained from CT, MRA, and electron microscopy modalities. The proposed method concatenates 3D image features extracted from three independent networks, a 3D CNN, and two time‐distributed ResNet BLSTM structures. The ultimate discriminative features are selected via the minimum redundancy maximum relevance (mRMR) feature selection method. Those features are then classified by a neural network model. Experiments adhere to the rules of the official splits and evaluation metrics of the MedMNIST3D datasets. The results reveal that the proposed approach outperforms similar studies in terms of accuracy and AUC.

The Suprascapular Artery as a Reference for the Evaluation of Supraspinatus Fatty Infiltration on Magnetic Resonance Imaging.

Fatty infiltration (FI) of the supraspinatus is commonly seen in chronic large-to-massive rotator cuff tears. Evaluating FI in patients with severe muscle atrophy can be confusing. The purpose of this study was to investigate the anatomic relationship between the suprascapular artery and supraspinatus muscle on magnetic resonance imaging (MRI) to provide a method for defining the border of the supraspinatus muscle on Y-view MRI. It was hypothesized that the branches of the suprascapular artery would encircle the supraspinatus muscle on Y-view and adjacent MRI slices and could be used for defining the supraspinatus outline on oblique sagittal images. Cohort study (diagnosis); Level of evidence, 3. A total of 172 shoulders that had undergone arthroscopic repair for large-to-massive rotator cuff tears were retrospectively included. Two methods, one based on the supraspinous fossa and trapezius (SF method) and the other based on the region bounded by the branches of the suprascapular artery (SA method), were used for defining the supraspinatus outline on Y-view MRI for the assessment of FI. Preoperative supraspinatus FI grade and tangent sign and postoperative tendon integrity were evaluated. Shoulder function was assessed using the American Shoulder and Elbow Surgeons (ASES) score and active range of motion. The branches of the suprascapular artery encircled the supraspinatus in all shoulders, with the diameter of the branches ranging from 0.5 to 3 mm. The agreement in supraspinatus FI grading between the SF and SA methods was 65.12%. When limited to the 61 shoulders with a positive tangent sign, the agreement dropped to 22.95% (κ = 0.032). The group (FI grade 2 by SA method and ≥3 by SF method) showed no significant difference in postoperative outcomes compared with the 2-vs-2 group but had significantly better postoperative ASES scores (P = .001) and active range of motion in forward elevation (P = .020) compared with the ≥3-vs-≥3 group. The tangent sign was positive in 92.16% of the 2-vs-≥3 group. The suprascapular artery is a reliable reference for the evaluation of supraspinatus FI. When a positive tangent sign presents, supraspinatus FI is likely to be overestimated if the hyperintensity outside the region bounded by the branches of the suprascapular artery is mistaken as FI.

2.5D deep learning based on multi-parameter MRI to differentiate primary lung cancer pathological subtypes in patients with brain metastases

BackgroundBrain metastases (BMs) represents a severe neurological complication stemming from cancers originating from various sources. It is a highly challenging clinical task to accurately distinguish the pathological subtypes of brain metastatic tumors from lung cancer (LC).The utility of 2.5-dimensional (2.5D) deep learning (DL) in distinguishing pathological subtypes of LC with BMs is yet to be determined. MethodsA total of 250 patients were included in this retrospective study, divided in a 7:3 ratio into training set (N=175) and testing set (N=75). We devised a method to assemble a series of two-dimensional (2D) images by extracting adjacent slices from a central slice in both superior-inferior and anterior-posterior directions to form a 2.5D dataset. Multi-Instance learning (MIL) is a weakly supervised learning method that organizes training instances into “bags” and provides labels for entire bags, with the purpose of learning a classifier based on the labeled positive and negative bags to predict the corresponding class for an unknown bag. Therefore, we employed MIL to construct a comprehensive 2.5D feature set. Then we used the single-slice as input for constructing the 2D model. DL features were extracted from these slices using the pre-trained ResNet101. All feature sets were inputted into the support vector machine (SVM) for evaluation. The diagnostic performance of the classification models were evaluated using five-fold cross-validation, with accuracy and area under the curve (AUC) metrics calculated for analysis. ResultsThe optimal performance was obtained using the 2.5D DL model, which achieved the micro-AUC of 0.868 (95% confidence interval [CI], 0.817–0.919) and accuracy of 0.836 in the test cohort. The 2D model achieved the micro-AUC of 0.836 (95 % CI, 0.778–0.894) and accuracy of 0.827 in the test cohort. ConclusionsThe proposed 2.5D DL model is feasible and effective in identifying pathological subtypes of BMs from lung cancer.

18F]Flotaza for Aβ Plaque Diagnostic Imaging: Evaluation in Postmortem Human Alzheimer's Disease Brain Hippocampus and PET/CT Imaging in 5xFAD Transgenic Mice.

The diagnostic value of imaging Aβ plaques in Alzheimer's disease (AD) has accelerated the development of fluorine-18 labeled radiotracers with a longer half-life for easier translation to clinical use. We have developed [18F]flotaza, which shows high binding to Aβ plaques in postmortem human AD brain slices with low white matter binding. We report the binding of [18F]flotaza in postmortem AD hippocampus compared to cognitively normal (CN) brains and the evaluation of [18F]flotaza in transgenic 5xFAD mice expressing Aβ plaques. [18F]Flotaza binding was assessed in well-characterized human postmortem brain tissue sections consisting of HP CA1-subiculum (HP CA1-SUB) regions in AD (n = 28; 13 male and 15 female) and CN subjects (n = 32; 16 male and 16 female). Adjacent slices were immunostained with anti-Aβ and analyzed using QuPath. In vitro and in vivo [18F]flotaza PET/CT studies were carried out in 5xFAD mice. Post-mortem human brain slices from all AD subjects were positively IHC stained with anti-Aβ. High [18F]flotaza binding was measured in the HP CA1-SUB grey matter (GM) regions compared to white matter (WM) of AD subjects with GM/WM > 100 in some subjects. The majority of CN subjects had no decipherable binding. Male AD exhibited greater WM than AD females (AD WM♂/WM♀ > 5; p < 0.001) but no difference amongst CN WM. In vitro studies in 5xFAD mice brain slices exhibited high binding [18F]flotaza ratios (>50 versus cerebellum) in the cortex, HP, and thalamus. In vivo, PET [18F]flotaza exhibited binding to Aβ plaques in 5xFAD mice with SUVR~1.4. [18F]Flotaza is a new Aβ plaque PET imaging agent that exhibited high binding to Aβ plaques in postmortem human AD. Along with the promising results in 5xFAD mice, the translation of [18F]flotaza to human PET studies may be worthwhile.

Fully and Weakly Supervised Deep Learning for Meniscal Injury Classification, and Location Based on MRI.

Meniscal injury is a common cause of knee joint pain and a precursor to knee osteoarthritis (KOA). The purpose of this study is to develop an automatic pipeline for meniscal injury classification and localization using fully and weakly supervised networks based on MRI images. In this retrospective study, data were from the osteoarthritis initiative (OAI). The MR images were reconstructed using a sagittal intermediate-weighted fat-suppressed turbo spin-echo sequence. (1) We used 130 knees from the OAI to develop the LGSA-UNet model which fuses the features of adjacent slices and adjusts the blocks in Siam to enable the central slice to obtain rich contextual information. (2) One thousand seven hundred and fifty-six knees from the OAI were included to establish segmentation and classification models. The segmentation model achieved a DICE coefficient ranging from 0.84 to 0.93. The AUC values ranged from 0.85 to 0.95 in the binary models. The accuracy for the three types of menisci (normal, tear, and maceration) ranged from 0.60 to 0.88. Furthermore, 206 knees from the orthopedic hospital were used as an external validation data set to evaluate the performance of the model. The segmentation and classification models still performed well on the external validation set. To compare the diagnostic performances between the deep learning (DL) models and radiologists, the external validation sets were sent to two radiologists. The binary classification model outperformed the diagnostic performance of the junior radiologist (0.82-0.87 versus 0.74-0.88). This study highlights the potential of DL in knee meniscus segmentation and injury classification which can help improve diagnostic efficiency.

Inversion of extinction coefficient for scattering media based on range-gated detection.

The extinction coefficient is an important parameter for describing light transmitting properties in the scattering medium. However, the single-point detection way and the large inversion error of the existing techniques cannot satisfy the need for spatial analysis and data application. A novel approach for inversion of extinction coefficient based on range-gated detection utilizing backward scattering of medium is proposed, and an inversion algorithm for extinction coefficient is established utilizing the backscattering signal of multiple adjacent spatial slices of the medium. The method can simultaneously invert transverse multiple-points extinction coefficient and longitudinal profile of extinction coefficient. Further, experiments are conducted in the scattering medium including fog and smoke based on the fabricated range-gated extinction-coefficient detection and inversion system, and the results demonstrate that inversion with the error less than 5% can be achieved at different detection distances, different concentrations and different kinds of scattering mediums. This approach offers a convenient, rapid and accurate means to acquire extinction coefficients, laying the foundation for the development of efficient environmental monitoring and high-quality defogging imaging.

Harmonizing three-dimensional MRI using pseudo-warping field guided GAN

In pursuit of cultivating automated models for magnetic resonance imaging (MRI) to aid in diagnostics, an escalating demand for extensive, multisite, and heterogeneous brain imaging datasets has emerged. This potentially introduces biased outcomes when directly applied for subsequent analysis. Researchers have endeavored to address this issue by pursuing the harmonization of MRIs. However, most existing image-based harmonization methods for MRI are tailored for 2D slices, which may introduce inter-slice variations when they are combined into a 3D volume. In this study, we aim to resolve inconsistencies between slices by introducing a pseudo-warping field. This field is created randomly and utilized to transform a slice into an artificially warped subsequent slice. The objective of this pseudo-warping field is to ensure that generators can consistently harmonize adjacent slices to another domain, without being affected by the varying content present in different slices. Furthermore, we construct unsupervised spatial and recycle loss to enhance the spatial accuracy and slice-wise consistency across the 3D images. The results demonstrate that our model effectively mitigates inter-slice variations and successfully preserves the anatomical details of the images during the harmonization process. Compared to generative harmonization models that employ 3D operators, our model exhibits greater computational efficiency and flexibility.

A novel quantitative method to evaluate lumbar disc degeneration: MRI histogram analysis.

This study aimed to use MRI histogram analysis to routine MRI sequences to evaluate lumbar disc degeneration (LDD), illustrate the correlation between this novel method and the traditional Pfirrmann classification method, and more importantly, perform comprehensive agreement analysis of MRI histogram analysis in various situations to evaluate its objectivity and stability. Lumbar MRI images from 133 subjects were included in this study. LDD was classified into grades by Pfirrmann classification and was measured as peak separation value by MRI histogram analysis. Correlation analysis between the two methods was performed and cutoff values were determined. In addition, the agreement analysis of peak separation value was performed by intraclass correlation coefficient (ICC) in four scenarios, including inter-resolution, inter-observer, inter-regions of interest (ROI) and inter-slice. Peak separation values were strongly correlated with Pfirrmann grades (r = - 0.847). The inter-resolution agreements of peak separation value between original image resolution of 2304 × 2304 and compressed image resolutions (1152 × 1152, 576 × 576, 288 × 288) were good to excellent (ICCs were 0.916, 0.876 and 0.822), except 144 × 144 was moderate (ICC = 533). The agreements of inter-observer (ICC = 0.982) and inter-ROI (ICC = 0.915) were excellent. Compared with the mid-sagittal slice, the inter-slice agreements were good for the first adjacent slices (ICCs were 0.826 and 0.844), and moderate to good for the second adjacent slices (ICC = 0.733 and 0.753). MRI histogram analysis, used in routine MRI sequences, demonstrated a strong correlation with Pfirrmann classification and good agreements in various scenarios, expanding the range of application and providing an effective, objective and quantitative tool to evaluate LDD.

Multiside graph neural network-based attention for local co-occurrence features fusion in lung nodule classification

Early diagnosis of lung cancer is critical as it can save people’s lives. Long-range dependencies within volumetric medical images are essential attributes for accurate lung nodule classification. Many deep learning-based methods are used for lung nodule classification; however, the construction of the lung nodule is not axial and can be any shape. Thus, the nodules are interrelated through their adjacent slices axially and diagonally, locally and globally, making their capture through local convolutional operations challenging. In this article, we benefited from local co-occurrences of texture features and graph neural networks (GNNs) to effectively capture important patterns considering the long-range dependencies among adjusted slices. The proposed framework comprises a multi-side graph construction layer (MSGCL) that computes informative texture features and captures spatial relationships from cross-sectional and longitude orientations of the nodule, creating two sets of nodes. Further, the graph-based fusion of long-range dependency layers (GFLL) is used to deeply fuse and generate attentive edges among fused nodes. The LIDC-IDRI dataset, used for training and testing the proposed Multi-side Graph Neural Network-based Attention for Local Co-occurrence Features Fusion (MS-GNN-ALCFF), achieves state-of-the-art. The LUNGx dataset used as an unseen dataset shows that our model is generalisable compared to baselines. The results of the proposed method when the LIDC-IDRI dataset was used for train and testing were 87.17 ± 0.84 %, 91.01 ± 1.14 %, 88.6 ± 1.42 %, 89.7 ± 0.87 %,95 ± 1.22 % and 76.7 ± 1.4 % in terms of Accuracy, Precision, Recall, F1-score, AUC and MCC respectively. When the LUNGx dataset is used for testing as an unseen dataset, the results were 69.86 ± 2.4 %, 75 ± 2.7 %, 71.4 ± 3.1 %, 73.17 ± 3.7 %, 70.2 ± 1.36 % and 71 ± 3.8 % for Accuracy, Precision, Recall, F1-score, AUC and MCC respectively. These findings represent the significance of the ability of GNNs to construct a multi-set of nodes through the proposed MSGCL layer and fuse deeply through the proposed GFLL layer.

Self-supervised anatomical continuity enhancement network for 7T SWI synthesis from 3T SWI

Synthesizing 7T Susceptibility Weighted Imaging (SWI) from 3T SWI could offer significant clinical benefits by combining the high sensitivity of 7T SWI for neurological disorders with the widespread availability of 3T SWI in diagnostic routines. Although methods exist for synthesizing 7T Magnetic Resonance Imaging (MRI), they primarily focus on traditional MRI modalities like T1-weighted imaging, rather than SWI. SWI poses unique challenges, including limited data availability and the invisibility of certain tissues in individual 3T SWI slices. To address these challenges, we propose a Self-supervised Anatomical Continuity Enhancement (SACE) network to synthesize 7T SWI from 3T SWI using plentiful 3T SWI data and limited 3T–7T paired data. The SACE employs two specifically designed pretext tasks to utilize low-level representations from abundant 3T SWI data for assisting 7T SWI synthesis in a downstream task with limited paired data. One pretext task emphasizes input-specific morphology by balancing the elimination of redundant patterns with the preservation of essential morphology, preventing the blurring of synthetic 7T SWI images. The other task improves the synthesis of tissues that are invisible in a single 3T SWI slice by aligning adjacent slices with the current slice and predicting their difference fields. The downstream task innovatively combines clinical knowledge with brain substructure diagrams to selectively enhance clinically relevant features. When evaluated on a dataset comprising 97 cases (5495 slices), the proposed method achieved a Peak Signal-to-Noise Ratio (PSNR) of 23.05 dB and a Structural Similarity Index (SSIM) of 0.688. Due to the absence of specific methods for 7T SWI, our method was compared with existing enhancement techniques for general 7T MRI synthesis, outperforming these techniques in the context of 7T SWI synthesis. Clinical evaluations have shown that our synthetic 7T SWI is clinically effective, demonstrating its potential as a clinical tool.

Deep learning-aided 3D proxy-bridged region-growing framework for multi-organ segmentation

Accurate multi-organ segmentation in 3D CT images is imperative for enhancing computer-aided diagnosis and radiotherapy planning. However, current deep learning-based methods for 3D multi-organ segmentation face challenges such as the need for labor-intensive manual pixel-level annotations and high hardware resource demands, especially regarding GPU resources. To address these issues, we propose a 3D proxy-bridged region-growing framework specifically designed for the segmentation of the liver and spleen. Specifically, a key slice is selected from each 3D volume according to the corresponding intensity histogram. Subsequently, a deep learning model is employed to pinpoint the semantic central patch on this key slice, to calculate the growing seed. To counteract the impact of noise, segmentation of the liver and spleen is conducted on superpixel images created through proxy-bridging strategy. The segmentation process is then extended to adjacent slices by applying the same methodology iteratively, culminating in the comprehensive segmentation results. Experimental results demonstrate that the proposed framework accomplishes segmentation of the liver and spleen with an average Dice Similarity Coefficient of approximately 0.93 and a Jaccard Similarity Coefficient of around 0.88. These outcomes substantiate the framework's capability to achieve performance on par with that of deep learning methods, albeit requiring less guidance information and lower GPU resources.

Anatomical Prior and Inter-slice Consistency for Semi-supervised Vertebral Structure Detection in 3D Ultrasound Volume.

Three-dimensional (3D) ultrasound imaging technique has been applied for scoliosis assessment, but the current assessment method only uses coronal projection images and cannot illustrate the 3D deformity and vertebra rotation. The vertebra detection is essential to reveal 3D spine information, but the detection task is challenging due to complex data and limited annotations. We propose VertMatch to detect vertebral structures in 3D ultrasound volume containing a detector and classifier. The detector network finds the potential positions of structures on transverse slice globally, and then the local patches are cropped based on detected positions. The classifier is used to distinguish whether the patches contain real vertebral structures and screen the predicted positions from the detector. VertMatch utilizes unlabeled data in a semi-supervised manner, and we develop two novel techniques for semi-supervised learning: (1) anatomical prior is used to acquire high-quality pseudo labels; (2) inter-slice consistency is used to utilize more unlabeled data by inputting multiple adjacent slices. Experimental results demonstrate that VertMatch can detect vertebra accurately in ultrasound volume and outperforms state-of-the-art methods. Moreover, VertMatch is also validated in automatic spinous process angle measurement on forty subjects with scoliosis, and the results illustrate that it can be a promising approach for the 3D assessment of scoliosis.

TTMRI: Multislice texture transformer network for undersampled MRI reconstruction

AbstractMagnetic resonance imaging (MRI) is a non‐interposition imaging technique that provides rich anatomical and physiological information. Yet it is limited by the long imaging time. Recently, deep neural networks have shown potential to significantly accelerate MRI. However, most of these approaches ignore the correlation between adjacent slices in MRI image sequences. In addition, the existing deep learning‐based methods for MRI are mainly based on convolutional neural networks (CNNs). They fail to capture long‐distance dependencies due to the small receptive field. Inspired by the feature similarity in adjacent slices and impressive performance of Transformer for exploiting the long‐distance dependencies, a novel multislice texture transformer network is presented for undersampled MRI reconstruction (TTMRI). Specifically, the proposed TTMRI is consisted of four modules, namely the texture extraction, correlation calculation, texture transfer and texture synthesis. It takes three adjacent slices as inputs, in which the middle one is the target image to be reconstructed, and the other two are auxiliary images. The multiscale features are extracted by the texture extraction module and their inter‐dependencies are calculated by the correlation calculation module, respectively. Then the relevant features are transferred by the texture transfer module and fused by the texture synthesis module. By considering inter‐slice correlations and leveraging the Transformer architecture, the joint feature learning across target and adjacent slices are encouraged. Moreover, TTMRI can be stacked with multiple layers to recover more texture information at different levels. Extensive experiments demonstrate that the proposed TTMRI outperforms other state‐of‐the‐art methods in both quantitative and qualitative evaluationsions.

NaMa: Neighbor-Aware Multi-Modal Adaptive Learning for Prostate Tumor Segmentation on Anisotropic MR Images

Accurate segmentation of prostate tumors from multi-modal magnetic resonance (MR) images is crucial for diagnosis and treatment of prostate cancer. However, the robustness of existing segmentation methods is limited, mainly because these methods 1) fail to adaptively assess subject-specific information of each MR modality for accurate tumor delineation, and 2) lack effective utilization of inter-slice information across thick slices in MR images to segment tumor as a whole 3D volume. In this work, we propose a two-stage neighbor-aware multi-modal adaptive learning network (NaMa) for accurate prostate tumor segmentation from multi-modal anisotropic MR images. In particular, in the first stage, we apply subject-specific multi-modal fusion in each slice by developing a novel modality-informativeness adaptive learning (MIAL) module for selecting and adaptively fusing informative representation of each modality based on inter-modality correlations. In the second stage, we exploit inter-slice feature correlations to derive volumetric tumor segmentation. Specifically, we first use a Unet variant with sequence layers to coarsely capture slice relationship at a global scale, and further generate an activation map for each slice. Then, we introduce an activation mapping guidance (AMG) module to refine slice-wise representation (via information from adjacent slices) for consistent tumor segmentation across neighboring slices. Besides, during the network training, we further apply a random mask strategy to each MR modality to improve feature representation efficiency. Experiments on both in-house and public (PICAI) multi-modal prostate tumor datasets show that our proposed NaMa performs better than state-of-the-art methods.

Abstract 1205: Inferring allele-specific copy number aberrations and tumor phylogeography from spatially resolved transcriptomics

Abstract Tumors evolve through the acquisition of somatic mutations over time and across space. While the temporal aspects of tumor evolution can be inferred by applying phylogenetic techniques to bulk or single-cell DNA sequencing data, the spatial aspects of tumor evolution remain understudied. Spatially resolved transcriptomics (SRT) measures gene expression at thousands of spatial locations in a tumor, but does not directly measure genetic aberrations. Thus, SRT data has not yet been fully utilized for tumor evolution studies. We introduce CalicoST, an algorithm that simultaneously infers allele-specific copy number aberrations (CNAs) and derives a phylogeographic model of tumor evolution from SRT data. By modeling CNA-induced perturbations in both gene expression and allele frequencies, CalicoST identifies important types of CNAs - including copy-neutral loss of heterozygosity (CNLOH) and mirrored subclonal CNAs - that are invisible to total copy number analysis. On SRT data from nine patients from the Human Tumor Atlas Network (HTAN), CalicoST achieves an average accuracy of 86%, approximately 21% higher than existing methods. On two of these patients with multiple adjacent slices, CalicoST reconstructs a tumor phylogeography that describes the spread of cancerous clones in three-dimensional space. CalicoST analysis of multiple SRT slices from a cancerous prostate organ reveals five spatially coherent clones, with mirrored subclonal CNAs distinguishing clones on the two sides of the prostate, and a bifurcating phylogeography in both genetic and physical space that suggests a polyclonal cancer. Citation Format: Cong Ma, Metin Balaban, Jingxian Liu, Siqi Chen, Li Ding, Benjamin J. Raphael. Inferring allele-specific copy number aberrations and tumor phylogeography from spatially resolved transcriptomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 1205.

Quantitative Evaluation Of Weld Defects Based On Overall Shape Three-Dimensional Reconstruction

The evaluation of the size of weld defects is crucial in assessing the quality of weld structures. In this study, a novel quantitative evaluation method for weld defects was proposed based on 3D reconstruction using slices. The method includes two significant contributions. First, the supervised defect segmentation method was introduced, which uses the results of the previous slice to supervise the segmentation of the current slice based on slight changes in adjacent defect slices. This ensures accurate segmentation of all defects. Second, the subpixel edge extraction method combining the Canny operator and cubic spline interpolation was proposed to improve the accuracy of edge detection. The proposed method was evaluated using 15 defects. The experimental results showed that the average errors of inclusion defects, incomplete-penetration defects, and incomplete-fusion defects were 13.6%, 8.18%, and 13.9%, respectively. Compared with the other methods, the proposed method not only had higher accuracy but also provided the volume value of defects.

PETS-Nets: Joint Pose Estimation and Tissue Segmentation of Fetal Brains Using Anatomy-Guided Networks.

Fetal Magnetic Resonance Imaging (MRI) is challenged by fetal movements and maternal breathing. Although fast MRI sequences allow artifact free acquisition of individual 2D slices, motion frequently occurs in the acquisition of spatially adjacent slices. Motion correction for each slice is thus critical for the reconstruction of 3D fetal brain MRI. In this paper, we propose a novel multi-task learning framework that adopts a coarse-to-fine strategy to jointly learn the pose estimation parameters for motion correction and tissue segmentation map of each slice in fetal MRI. Particularly, we design a regression-based segmentation loss as a deep supervision to learn anatomically more meaningful features for pose estimation and segmentation. In the coarse stage, a U-Net-like network learns the features shared for both tasks. In the refinement stage, to fully utilize the anatomical information, signed distance maps constructed from the coarse segmentation are introduced to guide the feature learning for both tasks. Finally, iterative incorporation of the signed distance maps further improves the performance of both regression and segmentation progressively. Experimental results of cross-validation across two different fetal datasets acquired with different scanners and imaging protocols demonstrate the effectiveness of the proposed method in reducing the pose estimation error and obtaining superior tissue segmentation results simultaneously, compared with state-of-the-art methods.

3D-CAM: a novel context-aware feature extraction framework for neurological disease classification.

In clinical practice and research, the classification and diagnosis of neurological diseases such as Parkinson's Disease (PD) and Multiple System Atrophy (MSA) have long posed a significant challenge. Currently, deep learning, as a cutting-edge technology, has demonstrated immense potential in computer-aided diagnosis of PD and MSA. However, existing methods rely heavily on manually selecting key feature slices and segmenting regions of interest. This not only increases subjectivity and complexity in the classification process but also limits the model's comprehensive analysis of global data features. To address this issue, this paper proposes a novel 3D context-aware modeling framework, named 3D-CAM. It considers 3D contextual information based on an attention mechanism. The framework, utilizing a 2D slicing-based strategy, innovatively integrates a Contextual Information Module and a Location Filtering Module. The Contextual Information Module can be applied to feature maps at any layer, effectively combining features from adjacent slices and utilizing an attention mechanism to focus on crucial features. The Location Filtering Module, on the other hand, is employed in the post-processing phase to filter significant slice segments of classification features. By employing this method in the fully automated classification of PD and MSA, an accuracy of 85.71%, a recall rate of 86.36%, and a precision of 90.48% were achieved. These results not only demonstrates potential for clinical applications, but also provides a novel perspective for medical image diagnosis, thereby offering robust support for accurate diagnosis of neurological diseases.