siRNADiscovery: a graph neural network for siRNA efficacy prediction via deep RNA sequence analysis.

The clinical adoption of small interfering RNAs (siRNAs) has prompted the development of various computational strategies for siRNA design, from traditional data analysis to advanced machine learning techniques. However, previous studies have inadequately considered the full complexity of the siRNA silencing mechanism, neglecting critical elements such as siRNA positioning on mRNA, RNA base-pairing probabilities, and RNA-AGO2 interactions, thereby limiting the insight and accuracy of existing models. Here, we introduce siRNADiscovery, a Graph Neural Network (GNN) framework that leverages both non-empirical and empirical rule-based features of siRNA and mRNA to effectively capture the complex dynamics of gene silencing. On multiple internal datasets, siRNADiscovery achieves state-of-the-art performance. Significantly, siRNADiscovery also outperforms existing methodologies in in vitro studies and on an externally validated dataset. Additionally, we develop a new data-splitting methodology that addresses the data leakage issue, a frequently overlooked problem in previous studies, ensuring the robustness and stability of our model under various experimental settings. Through rigorous testing, siRNADiscovery has demonstrated remarkable predictive accuracy and robustness, making significant contributions to the field of gene silencing. Furthermore, our approach to redefining data-splitting standards aims to set new benchmarks for future research in the domain of predictive biological modeling for siRNA.

Hepatic vessel segmentation based on 3D swin-transformer with inductive biased multi-head self-attention.

PURPOSE: Segmentation of liver vessels from CT images is indispensable prior to surgical planning and aroused a broad range of interest in the medical image analysis community. Due to the complex structure and low-contrast background, automatic liver vessel segmentation remains particularly challenging. Most of the related researches adopt FCN, U-net, and V-net variants as a backbone. However, these methods mainly focus on capturing multi-scale local features which may produce misclassified voxels due to the convolutional operator's limited locality reception field. METHODS: We propose a robust end-to-end vessel segmentation network called Inductive BIased Multi-Head Attention Vessel Net(IBIMHAV-Net) by expanding swin transformer to 3D and employing an effective combination of convolution and self-attention. In practice, we introduce voxel-wise embedding rather than patch-wise embedding to locate precise liver vessel voxels and adopt multi-scale convolutional operators to gain local spatial information. On the other hand, we propose the inductive biased multi-head self-attention which learns inductively biased relative positional embedding from initialized absolute position embedding. Based on this, we can gain more reliable queries and key matrices. RESULTS: We conducted experiments on the 3DIRCADb dataset. The average dice and sensitivity of the four tested cases were 74.8[Formula: see text] and 77.5[Formula: see text], which exceed the results of existing deep learning methods and improved graph cuts method. The Branches Detected(BD)/Tree-length Detected(TD) indexes also proved the global/local feature capture ability better than other methods. CONCLUSION: The proposed model IBIMHAV-Net provides an automatic, accurate 3D liver vessel segmentation with an interleaved architecture that better utilizes both global and local spatial features in CT volumes. It can be further extended for other clinical data.

Morphology-aware multi-source fusion-based intracranial aneurysms rupture prediction.

OBJECTIVES: We proposed a new approach to train deep learning model for aneurysm rupture prediction which only uses a limited amount of labeled data. METHOD: Using segmented aneurysm mask as input, a backbone model was pretrained using a self-supervised method to learn deep embeddings of aneurysm morphology from 947 unlabeled cases of angiographic images. Subsequently, the backbone model was finetuned using 120 labeled cases with known rupture status. Clinical information was integrated with deep embeddings to further improve prediction performance. The proposed model was compared with radiomics and conventional morphology models in prediction performance. An assistive diagnosis system was also developed based on the model and was tested with five neurosurgeons. RESULT: Our method achieved an area under the receiver operating characteristic curve (AUC) of 0.823, outperforming deep learning model trained from scratch (0.787). By integrating with clinical information, the proposed model's performance was further improved to AUC = 0.853, making the results significantly better than model based on radiomics (AUC = 0.805, p = 0.007) or model based on conventional morphology parameters (AUC = 0.766, p = 0.001). Our model also achieved the highest sensitivity, PPV, NPV, and accuracy among the others. Neurosurgeons' prediction performance was improved from AUC=0.877 to 0.945 (p = 0.037) with the assistive diagnosis system. CONCLUSION: Our proposed method could develop competitive deep learning model for rupture prediction using only a limited amount of data. The assistive diagnosis system could be useful for neurosurgeons to predict rupture. KEY POINTS: • A self-supervised learning method was proposed to mitigate the data-hungry issue of deep learning, enabling training deep neural network with a limited amount of data. • Using the proposed method, deep embeddings were extracted to represent intracranial aneurysm morphology. Prediction model based on deep embeddings was significantly better than conventional morphology model and radiomics model. • An assistive diagnosis system was developed using deep embeddings for case-based reasoning, which was shown to significantly improve neurosurgeons' performance to predict rupture.

Deep Learning for Automated Contouring of Primary Tumor Volumes by MRI for Nasopharyngeal Carcinoma.

Background Nasopharyngeal carcinoma (NPC) may be cured with radiation therapy. Tumor proximity to critical structures demands accuracy in tumor delineation to avoid toxicities from radiation therapy; however, tumor target contouring for head and neck radiation therapy is labor intensive and highly variable among radiation oncologists. Purpose To construct and validate an artificial intelligence (AI) contouring tool to automate primary gross tumor volume (GTV) contouring in patients with NPC. Materials and Methods In this retrospective study, MRI data sets covering the nasopharynx from 1021 patients (median age, 47 years; 751 male, 270 female) with NPC between September 2016 and September 2017 were collected and divided into training, validation, and testing cohorts of 715, 103, and 203 patients, respectively. GTV contours were delineated for 1021 patients and were defined by consensus of two experts. A three-dimensional convolutional neural network was applied to 818 training and validation MRI data sets to construct the AI tool, which was tested in 203 independent MRI data sets. Next, the AI tool was compared against eight qualified radiation oncologists in a multicenter evaluation by using a random sample of 20 test MRI examinations. The Wilcoxon matched-pairs signed rank test was used to compare the difference of Dice similarity coefficient (DSC) of pre- versus post-AI assistance. Results The AI-generated contours demonstrated a high level of accuracy when compared with ground truth contours at testing in 203 patients (DSC, 0.79; 2.0-mm difference in average surface distance). In multicenter evaluation, AI assistance improved contouring accuracy (five of eight oncologists had a higher median DSC after AI assistance; average median DSC, 0.74 vs 0.78; P < .001), reduced intra- and interobserver variation (by 36.4% and 54.5%, respectively), and reduced contouring time (by 39.4%). Conclusion The AI contouring tool improved primary gross tumor contouring accuracy of nasopharyngeal carcinoma, which could have a positive impact on tumor control and patient survival. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Chang in this issue.

Weakly supervised 3D deep learning for breast cancer classification and localization of the lesions in MR images.

BACKGROUND: The usefulness of 3D deep learning-based classification of breast cancer and malignancy localization from MRI has been reported. This work can potentially be very useful in the clinical domain and aid radiologists in breast cancer diagnosis. PURPOSE: To evaluate the efficacy of 3D deep convolutional neural network (CNN) for diagnosing breast cancer and localizing the lesions at dynamic contrast enhanced (DCE) MRI data in a weakly supervised manner. STUDY TYPE: Retrospective study. SUBJECTS: A total of 1537 female study cases (mean age 47.5 years ±11.8) were collected from March 2013 to December 2016. All the cases had labels of the pathology results as well as BI-RADS categories assessed by radiologists. FIELD STRENGTH/SEQUENCE: 1.5 T dynamic contrast-enhanced MRI. ASSESSMENT: Deep 3D densely connected networks were trained under image-level supervision to automatically classify the images and localize the lesions. The dataset was randomly divided into training (1073), validation (157), and testing (307) subsets. STATISTICAL TESTS: Accuracy, sensitivity, specificity, area under receiver operating characteristic curve (ROC), and the McNemar test for breast cancer classification. Dice similarity for breast cancer localization. RESULTS: The final algorithm performance for breast cancer diagnosis showed 83.7% (257 out of 307) accuracy (95% confidence interval [CI]: 79.1%, 87.4%), 90.8% (187 out of 206) sensitivity (95% CI: 80.6%, 94.1%), 69.3% (70 out of 101) specificity (95% CI: 59.7%, 77.5%), with the area under the curve ROC of 0.859. The weakly supervised cancer detection showed an overall Dice distance of 0.501 ± 0.274. DATA CONCLUSION: 3D CNNs demonstrated high accuracy for diagnosing breast cancer. The weakly supervised learning method showed promise for localizing lesions in volumetric radiology images with only image-level labels. LEVEL OF EVIDENCE: 4 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;50:1144-1151.

VoxResNet: Deep voxelwise residual networks for brain segmentation from 3D MR images.

Segmentation of key brain tissues from 3D medical images is of great significance for brain disease diagnosis, progression assessment and monitoring of neurologic conditions. While manual segmentation is time-consuming, laborious, and subjective, automated segmentation is quite challenging due to the complicated anatomical environment of brain and the large variations of brain tissues. We propose a novel voxelwise residual network (VoxResNet) with a set of effective training schemes to cope with this challenging problem. The main merit of residual learning is that it can alleviate the degradation problem when training a deep network so that the performance gains achieved by increasing the network depth can be fully leveraged. With this technique, our VoxResNet is built with 25 layers, and hence can generate more representative features to deal with the large variations of brain tissues than its rivals using hand-crafted features or shallower networks. In order to effectively train such a deep network with limited training data for brain segmentation, we seamlessly integrate multi-modality and multi-level contextual information into our network, so that the complementary information of different modalities can be harnessed and features of different scales can be exploited. Furthermore, an auto-context version of the VoxResNet is proposed by combining the low-level image appearance features, implicit shape information, and high-level context together for further improving the segmentation performance. Extensive experiments on the well-known benchmark (i.e., MRBrainS) of brain segmentation from 3D magnetic resonance (MR) images corroborated the efficacy of the proposed VoxResNet. Our method achieved the first place in the challenge out of 37 competitors including several state-of-the-art brain segmentation methods. Our method is inherently general and can be readily applied as a powerful tool to many brain-related studies, where accurate segmentation of brain structures is critical.

Personalized heterogeneous deformable model for fast volumetric registration.

BACKGROUND: Biomechanical deformable volumetric registration can help improve safety of surgical interventions by ensuring the operations are extremely precise. However, this technique has been limited by the accuracy and the computational efficiency of patient-specific modeling. METHODS: This study presents a tissue-tissue coupling strategy based on penalty method to model the heterogeneous behavior of deformable body, and estimate the personalized tissue-tissue coupling parameters in a data-driven way. Moreover, considering that the computational efficiency of biomechanical model is highly dependent on the mechanical resolution, a practical coarse-to-fine scheme is proposed to increase runtime efficiency. Particularly, a detail enrichment database is established in an offline fashion to represent the mapping relationship between the deformation results of high-resolution hexahedral mesh extracted from the raw medical data and a newly constructed low-resolution hexahedral mesh. At runtime, the mechanical behavior of human organ under interactions is simulated with this low-resolution hexahedral mesh, then the microstructures are synthesized in virtue of the detail enrichment database. RESULTS: The proposed method is validated by volumetric registration in an abdominal phantom compression experiments. Our personalized heterogeneous deformable model can well describe the coupling effects between different tissues of the phantom. Compared with high-resolution heterogeneous deformable model, the low-resolution deformable model with our detail enrichment database can achieve 9.4× faster, and the average target registration error is 3.42 mm, which demonstrates that the proposed method shows better volumetric registration performance than state-of-the-art. CONCLUSIONS: Our framework can well balance the precision and efficiency, and has great potential to be adopted in the practical augmented reality image-guided robotic systems.

Diagnostic Assessment of Deep Learning Algorithms for Detection of Lymph Node Metastases in Women With Breast Cancer.

Importance: Application of deep learning algorithms to whole-slide pathology images can potentially improve diagnostic accuracy and efficiency. Objective: Assess the performance of automated deep learning algorithms at detecting metastases in hematoxylin and eosin-stained tissue sections of lymph nodes of women with breast cancer and compare it with pathologists' diagnoses in a diagnostic setting. Design, Setting, and Participants: Researcher challenge competition (CAMELYON16) to develop automated solutions for detecting lymph node metastases (November 2015-November 2016). A training data set of whole-slide images from 2 centers in the Netherlands with (n = 110) and without (n = 160) nodal metastases verified by immunohistochemical staining were provided to challenge participants to build algorithms. Algorithm performance was evaluated in an independent test set of 129 whole-slide images (49 with and 80 without metastases). The same test set of corresponding glass slides was also evaluated by a panel of 11 pathologists with time constraint (WTC) from the Netherlands to ascertain likelihood of nodal metastases for each slide in a flexible 2-hour session, simulating routine pathology workflow, and by 1 pathologist without time constraint (WOTC). Exposures: Deep learning algorithms submitted as part of a challenge competition or pathologist interpretation. Main Outcomes and Measures: The presence of specific metastatic foci and the absence vs presence of lymph node metastasis in a slide or image using receiver operating characteristic curve analysis. The 11 pathologists participating in the simulation exercise rated their diagnostic confidence as definitely normal, probably normal, equivocal, probably tumor, or definitely tumor. Results: The area under the receiver operating characteristic curve (AUC) for the algorithms ranged from 0.556 to 0.994. The top-performing algorithm achieved a lesion-level, true-positive fraction comparable with that of the pathologist WOTC (72.4% [95% CI, 64.3%-80.4%]) at a mean of 0.0125 false-positives per normal whole-slide image. For the whole-slide image classification task, the best algorithm (AUC, 0.994 [95% CI, 0.983-0.999]) performed significantly better than the pathologists WTC in a diagnostic simulation (mean AUC, 0.810 [range, 0.738-0.884]; P < .001). The top 5 algorithms had a mean AUC that was comparable with the pathologist interpreting the slides in the absence of time constraints (mean AUC, 0.960 [range, 0.923-0.994] for the top 5 algorithms vs 0.966 [95% CI, 0.927-0.998] for the pathologist WOTC). Conclusions and Relevance: In the setting of a challenge competition, some deep learning algorithms achieved better diagnostic performance than a panel of 11 pathologists participating in a simulation exercise designed to mimic routine pathology workflow; algorithm performance was comparable with an expert pathologist interpreting whole-slide images without time constraints. Whether this approach has clinical utility will require evaluation in a clinical setting.

Automatic detection of arterial input function in dynamic contrast enhanced MRI based on affinity propagation clustering.

PURPOSE: To automatically and robustly detect the arterial input function (AIF) with high detection accuracy and low computational cost in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). MATERIALS AND METHODS: In this study, we developed an automatic AIF detection method using an accelerated version (Fast-AP) of affinity propagation (AP) clustering. The validity of this Fast-AP-based method was proved on two DCE-MRI datasets, i.e., rat kidney and human head and neck. The detailed AIF detection performance of this proposed method was assessed in comparison with other clustering-based methods, namely original AP and K-means, as well as the manual AIF detection method. RESULTS: Both the automatic AP- and Fast-AP-based methods achieved satisfactory AIF detection accuracy, but the computational cost of Fast-AP could be reduced by 64.37-92.10% on rat dataset and 73.18-90.18% on human dataset compared with the cost of AP. The K-means yielded the lowest computational cost, but resulted in the lowest AIF detection accuracy. The experimental results demonstrated that both the AP- and Fast-AP-based methods were insensitive to the initialization of cluster centers, and had superior robustness compared with K-means method. CONCLUSION: The Fast-AP-based method enables automatic AIF detection with high accuracy and efficiency.

A Dual Enrichment Synergistic Strategy to Handle Data Heterogeneity for Domain Incremental Cardiac Segmentation.

Upon remarkable progress in cardiac image segmentation, contemporary studies dedicate to further upgrading model functionality toward perfection, through progressively exploring the sequentially delivered datasets over time by domain incremental learning. Existing works mainly concentrated on addressing the heterogeneous style variations, but overlooked the critical shape variations across domains hidden behind the sub-disease composition discrepancy. In case the updated model catastrophically forgets the sub-diseases that were learned in past domains but are no longer present in the subsequent domains, we proposed a dual enrichment synergistic strategy to incrementally broaden model competence for a growing number of sub-diseases. The data-enriched scheme aims to diversify the shape composition of current training data via displacement-aware shape encoding and decoding, to gradually build up the robustness against cross-domain shape variations. Meanwhile, the model-enriched scheme intends to strengthen model capabilities by progressively appending and consolidating the latest expertise into a dynamically-expanded multi-expert network, to gradually cultivate the generalization ability over style-variated domains. The above two schemes work in synergy to collaboratively upgrade model capabilities in two-pronged manners. We have extensively evaluated our network with the ACDC and M&Ms datasets in single-domain and compound-domain incremental learning settings. Our approach outperformed other competing methods and achieved comparable results to the upper bound.

Effective Semi-Supervised Medical Image Segmentation with Probabilistic Representations and Prototype Learning.

Label scarcity, class imbalance and data uncertainty are three primary challenges that are commonly encountered in the semi-supervised medical image segmentation. In this work, we focus on the data uncertainty issue that is overlooked by previous literature. To address this issue, we propose a probabilistic prototype-based classifier that introduces uncertainty estimation into the entire pixel classification process, including probabilistic representation formulation, probabilistic pixel-prototype proximity matching, and distribution prototype update, leveraging principles from probability theory. By explicitly modeling data uncertainty at the pixel level, model robustness of our proposed framework to tricky pixels, such as ambiguous boundaries and noises, is greatly enhanced when compared to its deterministic counterpart and other uncertainty-aware strategy. Empirical evaluations on three publicly available datasets that exhibit severe boundary ambiguity show the superiority of our method over several competitors. Moreover, our method also demonstrates a stronger model robustness to simulated noisy data. Code is available at https://github.com/IsYuchenYuan/PPC.

Deep Omni-Supervised Learning for Rib Fracture Detection From Chest Radiology Images.

Deep learning (DL)-based rib fracture detection has shown promise of playing an important role in preventing mortality and improving patient outcome. Normally, developing DL-based object detection models requires a huge amount of bounding box annotation. However, annotating medical data is time-consuming and expertise-demanding, making obtaining a large amount of fine-grained annotations extremely infeasible. This poses a pressing need for developing label-efficient detection models to alleviate radiologists' labeling burden. To tackle this challenge, the literature on object detection has witnessed an increase of weakly-supervised and semi-supervised approaches, yet still lacks a unified framework that leverages various forms of fully-labeled, weakly-labeled, and unlabeled data. In this paper, we present a novel omni-supervised object detection network, ORF-Netv2, to leverage as much available supervision as possible. Specifically, a multi-branch omni-supervised detection head is introduced with each branch trained with a specific type of supervision. A co-training-based dynamic label assignment strategy is then proposed to enable flexible and robust learning from the weakly-labeled and unlabeled data. Extensive evaluation was conducted for the proposed framework with three rib fracture datasets on both chest CT and X-ray. By leveraging all forms of supervision, ORF-Netv2 achieves mAPs of 34.7, 44.7, and 19.4 on the three datasets, respectively, surpassing the baseline detector which uses only box annotations by mAP gains of 3.8, 4.8, and 5.0, respectively. Furthermore, ORF-Netv2 consistently outperforms other competitive label-efficient methods over various scenarios, showing a promising framework for label-efficient fracture detection. The code is available at: https://github.com/zhizhongchai/ORF-Net.

DC2T: Disentanglement-Guided Consolidation and Consistency Training for Semi-Supervised Cross-Site Continual Segmentation.

Continual Learning (CL) is recognized to be a storage-efficient and privacy-protecting approach for learning from sequentially-arriving medical sites. However, most existing CL methods assume that each site is fully labeled, which is impractical due to budget and expertise constraint. This paper studies the Semi-Supervised Continual Learning (SSCL) that adopts partially-labeled sites arriving over time, with each site delivering only limited labeled data while the majority remains unlabeled. In this regard, it is challenging to effectively utilize unlabeled data under dynamic cross-site domain gaps, leading to intractable model forgetting on such unlabeled data. To address this problem, we introduce a novel Disentanglement-guided Consolidation and Consistency Training (DC2T) framework, which roots in an Online Semi-Supervised representation Disentanglement (OSSD) perspective to excavate content representations of partially labeled data from sites arriving over time. Moreover, these content representations are required to be consolidated for site-invariance and calibrated for style-robustness, in order to alleviate forgetting even in the absence of ground truth. Specifically, for the invariance on previous sites, we retain historical content representations when learning on a new site, via a Content-inspired Parameter Consolidation (CPC) method that prevents altering the model parameters crucial for content preservation. For the robustness against style variation, we develop a Style-induced Consistency Training (SCT) scheme that enforces segmentation consistency over style-related perturbations to recalibrate content encoding. We extensively evaluate our method on fundus and cardiac image segmentation, indicating the advantage over existing SSCL methods for alleviating forgetting on unlabeled data.

Semantic-Oriented Visual Prompt Learning for Diabetic Retinopathy Grading on Fundus Images.

Diabetic retinopathy (DR) is a serious ocular condition that requires effective monitoring and treatment by ophthalmologists. However, constructing a reliable DR grading model remains a challenging and costly task, heavily reliant on high-quality training sets and adequate hardware resources. In this paper, we investigate the knowledge transferability of large-scale pre-trained models (LPMs) to fundus images based on prompt learning to construct a DR grading model efficiently. Unlike full-tuning which fine-tunes all parameters of LPMs, prompt learning only involves a minimal number of additional learnable parameters while achieving a competitive effect as full-tuning. Inspired by visual prompt tuning, we propose Semantic-oriented Visual Prompt Learning (SVPL) to enhance the semantic perception ability for better extracting task-specific knowledge from LPMs, without any additional annotations. Specifically, SVPL assigns a group of learnable prompts for each DR level to fit the complex pathological manifestations and then aligns each prompt group to task-specific semantic space via a contrastive group alignment (CGA) module. We also propose a plug-and-play adapter module, Hierarchical Semantic Delivery (HSD), which allows the semantic transition of prompt groups from shallow to deep layers to facilitate efficient knowledge mining and model convergence. Our extensive experiments on three public DR grading datasets demonstrate that SVPL achieves superior results compared to other transfer tuning and DR grading methods. Further analysis suggests that the generalized knowledge from LPMs is advantageous for constructing the DR grading model on fundus images.

3DSAM-adapter: Holistic adaptation of SAM from 2D to 3D for promptable tumor segmentation.

Despite that the segment anything model (SAM) achieved impressive results on general-purpose semantic segmentation with strong generalization ability on daily images, its demonstrated performance on medical image segmentation is less precise and unstable, especially when dealing with tumor segmentation tasks that involve objects of small sizes, irregular shapes, and low contrast. Notably, the original SAM architecture is designed for 2D natural images and, therefore would not be able to extract the 3D spatial information from volumetric medical data effectively. In this paper, we propose a novel adaptation method for transferring SAM from 2D to 3D for promptable medical image segmentation. Through a holistically designed scheme for architecture modification, we transfer the SAM to support volumetric inputs while retaining the majority of its pre-trained parameters for reuse. The fine-tuning process is conducted in a parameter-efficient manner, wherein most of the pre-trained parameters remain frozen, and only a few lightweight spatial adapters are introduced and tuned. Regardless of the domain gap between natural and medical data and the disparity in the spatial arrangement between 2D and 3D, the transformer trained on natural images can effectively capture the spatial patterns present in volumetric medical images with only lightweight adaptations. We conduct experiments on four open-source tumor segmentation datasets, and with a single click prompt, our model can outperform domain state-of-the-art medical image segmentation models and interactive segmentation models. We also compared our adaptation method with existing popular adapters and observed significant performance improvement on most datasets. Our code and models are available at: https://github.com/med-air/3DSAM-adapter.

Scale-Aware Super-Resolution Network With Dual Affinity Learning for Lesion Segmentation From Medical Images.

Convolutional neural networks (CNNs) have shown remarkable progress in medical image segmentation. However, the lesion segmentation remains a challenge to state-of-the-art CNN-based algorithms due to the variance in scales and shapes. On the one hand, tiny lesions are hard to delineate precisely from the medical images which are often of low resolutions. On the other hand, segmenting large-size lesions requires large receptive fields, which exacerbates the first challenge. In this article, we present a scale-aware super-resolution (SR) network to adaptively segment lesions of various sizes from low-resolution (LR) medical images. Our proposed network contains dual branches to simultaneously conduct lesion mask SR (LMSR) and lesion image SR (LISR). Meanwhile, we introduce scale-aware dilated convolution (SDC) blocks into the multitask decoders to adaptively adjust the receptive fields of the convolutional kernels according to the lesion sizes. To guide the segmentation branch to learn from richer high-resolution (HR) features, we propose a feature affinity (FA) module and a scale affinity (SA) module to enhance the multitask learning of the dual branches. On multiple challenging lesion segmentation datasets, our proposed network achieved consistent improvements compared with other state-of-the-art methods. Code will be available at: https://github.com/poiuohke/SASR_Net.

Video Instance Shadow Detection Under the Sun and Sky.

Instance shadow detection, crucial for applications such as photo editing and light direction estimation, has undergone significant advancements in predicting shadow instances, object instances, and their associations. The extension of this task to videos presents challenges in annotating diverse video data and addressing complexities arising from occlusion and temporary disappearances within associations. In response to these challenges, we introduce ViShadow, a semi-supervised video instance shadow detection framework that leverages both labeled image data and unlabeled video data for training. ViShadow features a two-stage training pipeline: the first stage, utilizing labeled image data, identifies shadow and object instances through contrastive learning for cross-frame pairing. The second stage employs unlabeled videos, incorporating an associated cycle consistency loss to enhance tracking ability. A retrieval mechanism is introduced to manage temporary disappearances, ensuring tracking continuity. The SOBA-VID dataset, comprising unlabeled training videos and labeled testing videos, along with the SOAP-VID metric, is introduced for the quantitative evaluation of VISD solutions. The effectiveness of ViShadow is further demonstrated through various video-level applications such as video inpainting, instance cloning, shadow editing, and text-instructed shadow-object manipulation.

S2Former-OR: Single-Stage Bi-Modal Transformer for Scene Graph Generation in OR.

Scene graph generation (SGG) of surgical procedures is crucial in enhancing holistically cognitive intelligence in the operating room (OR). However, previous works have primarily relied on multi-stage learning, where the generated semantic scene graphs depend on intermediate processes with pose estimation and object detection. This pipeline may potentially compromise the flexibility of learning multimodal representations, consequently constraining the overall effectiveness. In this study, we introduce a novel single-stage bi-modal transformer framework for SGG in the OR, termed, S2Former-OR, aimed to complementally leverage multi-view 2D scenes and 3D point clouds for SGG in an end-to-end manner. Concretely, our model embraces a View-Sync Transfusion scheme to encourage multi-view visual information interaction. Concurrently, a Geometry-Visual Cohesion operation is designed to integrate the synergic 2D semantic features into 3D point cloud features. Moreover, based on the augmented feature, we propose a novel relation-sensitive transformer decoder that embeds dynamic entity-pair queries and relational trait priors, which enables the direct prediction of entity-pair relations for graph generation without intermediate steps. Extensive experiments have validated the superior SGG performance and lower computational cost of S2Former-OR on 4D-OR benchmark, compared with current OR-SGG methods, e.g., 3 percentage points Precision increase and 24.2M reduction in model parameters. We further compared our method with generic single-stage SGG methods with broader metrics for a comprehensive evaluation, with consistently better performance achieved. Our source code can be made available at: https://github.com/PJLallen/S2Former-OR.

CalibNet: Dual-Branch Cross-Modal Calibration for RGB-D Salient Instance Segmentation.

In this study, we propose a novel approach for RGB-D salient instance segmentation using a dual-branch cross-modal feature calibration architecture called CalibNet. Our method simultaneously calibrates depth and RGB features in the kernel and mask branches to generate instance-aware kernels and mask features. CalibNet consists of three simple modules, a dynamic interactive kernel (DIK) and a weight-sharing fusion (WSF), which work together to generate effective instance-aware kernels and integrate cross-modal features. To improve the quality of depth features, we incorporate a depth similarity assessment (DSA) module prior to DIK and WSF. In addition, we further contribute a new DSIS dataset, which contains 1,940 images with elaborate instance-level annotations. Extensive experiments on three challenging benchmarks show that CalibNet yields a promising result, i.e., 58.0% AP with 320×480 input size on the COME15K-E test set, which significantly surpasses the alternative frameworks. Our code and dataset will be publicly available at: https://github.com/PJLallen/CalibNet.

Enabling target-aware molecule generation to follow multi objectives with Pareto MCTS.

Target-aware drug discovery has greatly accelerated the drug discovery process to design small-molecule ligands with high binding affinity to disease-related protein targets. Conditioned on targeted proteins, previous works utilize various kinds of deep generative models and have shown great potential in generating molecules with strong protein-ligand binding interactions. However, beyond binding affinity, effective drug molecules must manifest other essential properties such as high drug-likeness, which are not explicitly addressed by current target-aware generative methods. In this article, aiming to bridge the gap of multi-objective target-aware molecule generation in the field of deep learning-based drug discovery, we propose ParetoDrug, a Pareto Monte Carlo Tree Search (MCTS) generation algorithm. ParetoDrug searches molecules on the Pareto Front in chemical space using MCTS to enable synchronous optimization of multiple properties. Specifically, ParetoDrug utilizes pretrained atom-by-atom autoregressive generative models for the exploration guidance to desired molecules during MCTS searching. Besides, when selecting the next atom symbol, a scheme named ParetoPUCT is proposed to balance exploration and exploitation. Benchmark experiments and case studies demonstrate that ParetoDrug is highly effective in traversing the large and complex chemical space to discover novel compounds with satisfactory binding affinities and drug-like properties for various multi-objective target-aware drug discovery tasks.