Learning multi-label image recognition with incomplete annotation is gaining popularity due to its superior performance and significant labor savings when compared to training with fully labeled datasets. Existing literature mainly focuses on label completion and co-occurrence learning while facing difficulties with the most common single-positive label manner. To tackle this problem, we present a semantic contrastive bootstrapping (Scob) approach to gradually recover the cross-object relationships by introducing class activation as semantic guidance. With this learning guidance, we then propose a recurrent semantic masked transformer to extract iconic object-level representations and delve into the contrastive learning problems on multi-label classification tasks. We further propose a bootstrapping framework in an Expectation-Maximization fashion that iteratively optimizes the network parameters and refines semantic guidance to alleviate possible disturbance caused by wrong semantic guidance. Extensive experimental results demonstrate that the proposed joint learning framework surpasses the state-of-the-art models by a large margin on four public multi-label image recognition benchmarks. Codes can be found at https://github.com/iCVTEAM/Scob.
Automatic analysis of colonoscopy images has been an active field of research motivated by the importance of early detection of precancerous polyps. However, detecting polyps during the live examination can be challenging due to various factors such as variation of skills and experience among the endoscopists, lack of attentiveness, and fatigue leading to a high polyp miss-rate. Deep learning has emerged as a promising solution to this challenge as it can assist endoscopists in detecting and classifying overlooked polyps and abnormalities in real time. In addition to the algorithm's accuracy, transparency and interpretability are crucial to explaining the whys and hows of the algorithm's prediction. Further, most algorithms are developed in private data, closed source, or proprietary software, and methods lack reproducibility. Therefore, to promote the development of efficient and transparent methods, we have organized the "Medico automatic polyp segmentation (Medico 2020)" and "MedAI: Transparency in Medical Image Segmentation (MedAI 2021)" competitions. We present a comprehensive summary and analyze each contribution, highlight the strength of the best-performing methods, and discuss the possibility of clinical translations of such methods into the clinic. For the transparency task, a multi-disciplinary team, including expert gastroenterologists, accessed each submission and evaluated the team based on open-source practices, failure case analysis, ablation studies, usability and understandability of evaluations to gain a deeper understanding of the models' credibility for clinical deployment. Through the comprehensive analysis of the challenge, we not only highlight the advancements in polyp and surgical instrument segmentation but also encourage qualitative evaluation for building more transparent and understandable AI-based colonoscopy systems.
Markerless motion capture (MMC) is revolutionizing gait analysis in clinical settings by making it more accessible, raising the question of how to extract the most clinically meaningful information from gait data. In multiple fields ranging from image processing to natural language processing, self-supervised learning (SSL) from large amounts of unannotated data produces very effective representations for downstream tasks. However, there has only been limited use of SSL to learn effective representations of gait and movement, and it has not been applied to gait analysis with MMC. One SSL objective that has not been applied to gait is contrastive learning, which finds representations that place similar samples closer together in the learned space. If the learned similarity metric captures clinically meaningful differences, this could produce a useful representation for many downstream clinical tasks. Contrastive learning can also be combined with causal masking to predict future timesteps, which is an appealing SSL objective given the dynamical nature of gait. We applied these techniques to gait analyses performed with MMC in a rehabilitation hospital from a diverse clinical population. We find that contrastive learning on unannotated gait data learns a representation that captures clinically meaningful information. We probe this learned representation using the framework of biomarkers and show it holds promise as both a diagnostic and response biomarker, by showing it can accurately classify diagnosis from gait and is responsive to inpatient therapy, respectively. We ultimately hope these learned representations will enable predictive and prognostic gait-based biomarkers that can facilitate precision rehabilitation through greater use of MMC to quantify movement in rehabilitation.
Neural network prediction probabilities and accuracy are often only weakly-correlated. Inherent label ambiguity in training data for image segmentation aggravates such miscalibration. We show that logit consistency across stochastic transformations acts as a spatially varying regularizer that prevents overconfident predictions at pixels with ambiguous labels. Our boundary-weighted extension of this regularizer provides state-of-the-art calibration for prostate and heart MRI segmentation.
Training a deep learning model to classify histopathological images is challenging, because of the color and shape variability of the cells and tissues, and the reduced amount of available data, which does not allow proper learning of those variations. Variations can come from the image acquisition process, for example, due to different cell staining protocols or tissue deformation. To tackle this challenge, Data Augmentation (DA) can be used during training to generate additional samples by applying transformations to existing ones, to help the model become invariant to those color and shape transformations. The problem with DA is that it is not only dataset-specific but it also requires domain knowledge, which is not always available. Without this knowledge, selecting the right transformations can only be done using heuristics or through a computationally demanding search. To address this, we propose an automatic DA learning method. In this method, the DA parameters, i.e. the transformation parameters needed to improve the model training, are considered learnable and are learned automatically using a bilevel optimization approach in a quick and efficient way using truncated backpropagation. We validated the method on six different datasets. Experimental results show that our model can learn color and affine transformations that are more helpful to train an image classifier than predefined DA transformations, which are also more expensive as they need to be selected before the training by grid search on a validation set. We also show that similarly to a model trained with RandAugment, our model has also only a few method-specific hyperparameters to tune but is performing better. This makes our model a good solution for learning the best DA parameters, especially in the context of histopathological images, where defining potentially useful transformation heuristically is not trivial.
Recent advances in deep learning-based medical image segmentation studies achieve nearly human-level performance in fully supervised manner. However, acquiring pixel-level expert annotations is extremely expensive and laborious in medical imaging fields. Unsupervised domain adaptation (UDA) can alleviate this problem, which makes it possible to use annotated data in one imaging modality to train a network that can successfully perform segmentation on target imaging modality with no labels. In this work, we propose SDC-UDA, a simple yet effective volumetric UDA framework for slice-direction continuous cross-modality medical image segmentation which combines intra- and inter-slice self-attentive image translation, uncertainty-constrained pseudo-label refinement, and volumetric self-training. Our method is distinguished from previous methods on UDA for medical image segmentation in that it can obtain continuous segmentation in the slice direction, thereby ensuring higher accuracy and potential in clinical practice. We validate SDC-UDA with multiple publicly available cross-modality medical image segmentation datasets and achieve state-of-the-art segmentation performance, not to mention the superior slice-direction continuity of prediction compared to previous studies.
Accurate segmentation of fetal brain magnetic resonance images is crucial for analyzing fetal brain development and detecting potential neurodevelopmental abnormalities. Traditional deep learning-based automatic segmentation, although effective, requires extensive training data with ground-truth labels, typically produced by clinicians through a time-consuming annotation process. To overcome this challenge, we propose a novel unsupervised segmentation method based on multi-atlas segmentation, that accurately segments multiple tissues without relying on labeled data for training. Our method employs a cascaded deep learning network for 3D image registration, which computes small, incremental deformations to the moving image to align it precisely with the fixed image. This cascaded network can then be used to register multiple annotated images with the image to be segmented, and combine the propagated labels to form a refined segmentation. Our experiments demonstrate that the proposed cascaded architecture outperforms the state-of-the-art registration methods that were tested. Furthermore, the derived segmentation method achieves similar performance and inference time to nnU-Net while only using a small subset of annotated data for the multi-atlas segmentation task and none for training the network. Our pipeline for registration and multi-atlas segmentation is publicly available at https://github.com/ValBcn/CasReg.
With an enormous number of hand images generated over time, unleashing pose knowledge from unlabeled images for supervised hand mesh estimation is an emerging yet challenging topic. To alleviate this issue, semi-supervised and self-supervised approaches have been proposed, but they are limited by the reliance on detection models or conventional ResNet backbones. In this paper, inspired by the rapid progress of Masked Image Modeling (MIM) in visual classification tasks, we propose a novel self-supervised pre-training strategy for regressing 3D hand mesh parameters. Our approach involves a unified and multi-granularity strategy that includes a pseudo keypoint alignment module in the teacher-student framework for learning pose-aware semantic class tokens. For patch tokens with detailed locality, we adopt a self-distillation manner between teacher and student network based on MIM pre-training. To better fit low-level regression tasks, we incorporate pixel reconstruction tasks for multi-level representation learning. Additionally, we design a strong pose estimation baseline using a simple vanilla vision Transformer (ViT) as the backbone and attach a PyMAF head after tokens for regression. Extensive experiments demonstrate that our proposed approach, named HandMIM, achieves strong performance on various hand mesh estimation tasks. Notably, HandMIM outperforms specially optimized architectures, achieving 6.29mm and 8.00mm PAVPE (Vertex-Point-Error) on challenging FreiHAND and HO3Dv2 test sets, respectively, establishing new state-of-the-art records on 3D hand mesh estimation.
Whole-body pose estimation localizes the human body, hand, face, and foot keypoints in an image. This task is challenging due to multi-scale body parts, fine-grained localization for low-resolution regions, and data scarcity. Meanwhile, applying a highly efficient and accurate pose estimator to widely human-centric understanding and generation tasks is urgent. In this work, we present a two-stage pose \textbf{D}istillation for \textbf{W}hole-body \textbf{P}ose estimators, named \textbf{DWPose}, to improve their effectiveness and efficiency. The first-stage distillation designs a weight-decay strategy while utilizing a teacher's intermediate feature and final logits with both visible and invisible keypoints to supervise the student from scratch. The second stage distills the student model itself to further improve performance. Different from the previous self-knowledge distillation, this stage finetunes the student's head with only 20% training time as a plug-and-play training strategy. For data limitations, we explore the UBody dataset that contains diverse facial expressions and hand gestures for real-life applications. Comprehensive experiments show the superiority of our proposed simple yet effective methods. We achieve new state-of-the-art performance on COCO-WholeBody, significantly boosting the whole-body AP of RTMPose-l from 64.8% to 66.5%, even surpassing RTMPose-x teacher with 65.3% AP. We release a series of models with different sizes, from tiny to large, for satisfying various downstream tasks. Our codes and models are available at https://github.com/IDEA-Research/DWPose.
The essence of self-supervised image denoising is to restore the signal from the noisy image alone. State-of-the-art solutions for this task rely on the idea of masking pixels and training a fully-convolutional neural network to impute them. This most often requires multiple forward passes, information about the noise model, and intricate regularization functions. In this paper, we propose a Swin Transformer-based Image Autoencoder (SwinIA), the first convolution-free architecture for self-supervised denoising. It can be trained end-to-end with a simple mean squared error loss without masking and does not require any prior knowledge about clean data or noise distribution. Despite its simplicity, SwinIA establishes state-of-the-art on several common benchmarks.