Search

Machine learning (ML) models have significantly impacted various domains in our everyday lives. While large language models (LLMs) offer intuitive interfaces and versatility, task-specific ML models remain valuable for their efficiency and focused performance in specialized tasks. However, developing these models requires technical expertise, making it particularly challenging for non-expert users to customize them for their unique needs. Although interactive machine learning (IML) aims to democratize ML development through user-friendly interfaces, users struggle to translate their requirements into appropriate ML tasks. We propose human-LLM collaborative ML as a new paradigm bridging human-driven IML and machine-driven LLM approaches. To realize this vision, we introduce DuetML, a framework that integrates multimodal LLMs (MLLMs) as interactive agents collaborating with users throughout the ML process. Our system carefully balances MLLM capabilities with user agency by implementing both reactive and proactive interactions between users and MLLM agents. Through a comparative user study, we demonstrate that DuetML enables non-expert users to define training data that better aligns with target tasks without increasing cognitive load, while offering opportunities for deeper engagement with ML task formulation., Comment: 22 pages, 10 figures

Radiography is widely used in orthopedics for its affordability and low radiation exposure. 3D reconstruction from a single radiograph, so-called 2D-3D reconstruction, offers the possibility of various clinical applications, but achieving clinically viable accuracy and computational efficiency is still an unsolved challenge. Unlike other areas in computer vision, X-ray imaging's unique properties, such as ray penetration and fixed geometry, have not been fully exploited. We propose a novel approach that simultaneously learns multiple depth maps (front- and back-surface of multiple bones) derived from the X-ray image to computed tomography registration. The proposed method not only leverages the fixed geometry characteristic of X-ray imaging but also enhances the precision of the reconstruction of the whole surface. Our study involved 600 CT and 2651 X-ray images (4 to 5 posed X-ray images per patient), demonstrating our method's superiority over traditional approaches with a surface reconstruction error reduction from 4.78 mm to 1.96 mm. This significant accuracy improvement and enhanced computational efficiency suggest our approach's potential for clinical application., Comment: MICCAI 2024. 12 pages, 4 figures

Deep learning-based image segmentation has allowed for the fully automated, accurate, and rapid analysis of musculoskeletal (MSK) structures from medical images. However, current approaches were either applied only to 2D cross-sectional images, addressed few structures, or were validated on small datasets, which limit the application in large-scale databases. This study aimed to validate an improved deep learning model for volumetric MSK segmentation of the hip and thigh with uncertainty estimation from clinical computed tomography (CT) images. Databases of CT images from multiple manufacturers/scanners, disease status, and patient positioning were used. The segmentation accuracy, and accuracy in estimating the structures volume and density, i.e., mean HU, were evaluated. An approach for segmentation failure detection based on predictive uncertainty was also investigated. The model has shown an overall improvement with respect to all segmentation accuracy and structure volume/density evaluation metrics. The predictive uncertainty yielded large areas under the receiver operating characteristic (AUROC) curves (AUROCs>=.95) in detecting inaccurate and failed segmentations. The high segmentation and muscle volume/density estimation accuracy, along with the high accuracy in failure detection based on the predictive uncertainty, exhibited the model's reliability for analyzing individual MSK structures in large-scale CT databases., Comment: 29 pages, 7+10supp figures, 8 tables

While most vision tasks are essentially visual in nature (for recognition), some important tasks, especially in the medical field, also require quantitative analysis (for quantification) using quantitative images. Unlike in visual analysis, pixel values in quantitative images correspond to physical metrics measured by specific devices (e.g., a depth image). However, recent work has shown that it is sometimes possible to synthesize accurate quantitative values from visual ones (e.g., depth from visual cues or defocus). This research aims to improve quantitative image synthesis (QIS) by exploring pretraining and image resolution scaling. We propose a benchmark for evaluating pretraining performance using the task of QIS-based bone mineral density (BMD) estimation from plain X-ray images, where the synthesized quantitative image is used to derive BMD. Our results show that appropriate pretraining can improve QIS performance, significantly raising the correlation of BMD estimation from 0.820 to 0.898, while others do not help or even hinder it. Scaling-up the resolution can further boost the correlation up to 0.923, a significant enhancement over conventional methods. Future work will include exploring more pretraining strategies and validating them on other image synthesis tasks., Comment: SASHIMI, 2024 (MICCAI workshop). 13 pages, 3 figures

A versatile robot working in a domestic environment based on a deep neural network (DNN) is currently attracting attention. One of the roles expected for domestic robots is caregiving for a human. In particular, we focus on repositioning care because repositioning plays a fundamental role in supporting the health and quality of life of individuals with limited mobility. However, generating motions of the repositioning care, avoiding applying force to non-target parts and applying appropriate force to target parts, remains challenging. In this study, we proposed a DNN-based architecture using visual and somatosensory attention mechanisms that can generate dual-arm repositioning motions which involve different sequential policies of interaction force; contact-less reaching and contact-based assisting motions. We used the humanoid robot Dry-AIREC, which features the capability to adjust joint impedance dynamically. In the experiment, the repositioning assistance from the supine position to the sitting position was conducted by Dry-AIREC. The trained model, utilizing the proposed architecture, successfully guided the robot's hand to the back of the mannequin without excessive contact force on the mannequin and provided adequate support and appropriate contact for postural adjustment.

We highlight that match fixed effects, represented by the coefficients of interaction terms involving dummy variables for two elements, lack identification without specific restrictions on parameters. Consequently, the coefficients typically reported as relative match fixed effects by statistical software are not interpretable. To address this, we establish normalization conditions that enable identification of match fixed effect parameters as interpretable indicators of unobserved match affinity, facilitating comparisons among observed matches. Using data from middle school students in the 2007 Trends in International Mathematics and Science Study (TIMSS), we highlight the distribution of comparable match fixed effects within a specific school., Comment: 6 pages, 4 page appendix

The development of algorithms to accurately decode of neural information is a long-standing effort in the field of neuroscience. Brain decoding is typically employed by training machine learning models to map neural data onto a preestablished vector representation of stimulus features. These vectors are usually derived from image- and/or text-based feature spaces. Nonetheless, the intrinsic characteristics of these vectors might be fundamentally different than those encoded by the brain, limiting the ability of algorithms to accurately learn this mapping. To address this issue, here, we propose a representation learning framework, called brain-aligning of semantic vectors, that fine-tunes pretrained feature vectors to better align with the structure of neural representations of visual stimuli in the human brain. We trained this model with functional magnetic resonance imaging (fMRI) data representing 150 visual stimulus categories; then, we performed zero-shot brain decoding on 1) fMRI, 2) magnetoencephalography (MEG), and 3) electrocorticography (ECoG) data reflecting neural representations of visual stimuli. By using fMRI-based brain-aligned vectors, the zero-shot decoding accuracy all three neuroimaging datasets increased. This finding underscores the potential of leveraging a richer array of brainderived features to increase the performance of brain decoding algorithms., Comment: 40 pages, 5 figures

Appearance-based gaze estimation, which uses only a regular camera to estimate human gaze, is important in various application fields. While the technique faces data bias issues, data collection protocol is often demanding, and collecting data from a wide range of participants is difficult. It is an important challenge to design opportunities that allow a diverse range of people to participate while ensuring the quality of the training data. To tackle this challenge, we introduce a novel gamified approach for collecting training data. In this game, two players communicate words via eye gaze through a transparent letter board. Images captured during gameplay serve as valuable training data for gaze estimation models. The game is designed as a physical installation that involves communication between players, and it is expected to attract the interest of diverse participants. We assess the game's significance on data quality and user experience through a comparative user study., Comment: Accepted to International Journal of Human-Computer Interaction (IJHCI)

Progression of hip osteoarthritis (hip OA) leads to pain and disability, likely leading to surgical treatment such as hip arthroplasty at the terminal stage. The severity of hip OA is often classified using the Crowe and Kellgren-Lawrence (KL) classifications. However, as the classification is subjective, we aimed to develop an automated approach to classify the disease severity based on the two grades using digitally-reconstructed radiographs (DRRs) from CT images. Automatic grading of the hip OA severity was performed using deep learning-based models. The models were trained to predict the disease grade using two grading schemes, i.e., predicting the Crowe and KL grades separately, and predicting a new ordinal label combining both grades and representing the disease progression of hip OA. The models were trained in classification and regression settings. In addition, the model uncertainty was estimated and validated as a predictor of classification accuracy. The models were trained and validated on a database of 197 hip OA patients, and externally validated on 52 patients. The model accuracy was evaluated using exact class accuracy (ECA), one-neighbor class accuracy (ONCA), and balanced accuracy.The deep learning models produced a comparable accuracy of approximately 0.65 (ECA) and 0.95 (ONCA) in the classification and regression settings. The model uncertainty was significantly larger in cases with large classification errors (P<6e-3). In this study, an automatic approach for grading hip OA severity from CT images was developed. The models have shown comparable performance with high ONCA, which facilitates automated grading in large-scale CT databases and indicates the potential for further disease progression analysis. Classification accuracy was correlated with the model uncertainty, which would allow for the prediction of classification errors.

We introduce StreamDiffusion, a real-time diffusion pipeline designed for interactive image generation. Existing diffusion models are adept at creating images from text or image prompts, yet they often fall short in real-time interaction. This limitation becomes particularly evident in scenarios involving continuous input, such as Metaverse, live video streaming, and broadcasting, where high throughput is imperative. To address this, we present a novel approach that transforms the original sequential denoising into the batching denoising process. Stream Batch eliminates the conventional wait-and-interact approach and enables fluid and high throughput streams. To handle the frequency disparity between data input and model throughput, we design a novel input-output queue for parallelizing the streaming process. Moreover, the existing diffusion pipeline uses classifier-free guidance(CFG), which requires additional U-Net computation. To mitigate the redundant computations, we propose a novel residual classifier-free guidance (RCFG) algorithm that reduces the number of negative conditional denoising steps to only one or even zero. Besides, we introduce a stochastic similarity filter(SSF) to optimize power consumption. Our Stream Batch achieves around 1.5x speedup compared to the sequential denoising method at different denoising levels. The proposed RCFG leads to speeds up to 2.05x higher than the conventional CFG. Combining the proposed strategies and existing mature acceleration tools makes the image-to-image generation achieve up-to 91.07fps on one RTX4090, improving the throughputs of AutoPipline developed by Diffusers over 59.56x. Furthermore, our proposed StreamDiffusion also significantly reduces the energy consumption by 2.39x on one RTX3060 and 1.99x on one RTX4090, respectively., Comment: tech report, the code is available at https://github.com/cumulo-autumn/StreamDiffusion

Skill transfer from humans to robots is challenging. Presently, many researchers focus on capturing only position or joint angle data from humans to teach the robots. Even though this approach has yielded impressive results for grasping applications, reconstructing motion for object handling or fine manipulation from a human hand to a robot hand has been sparsely explored. Humans use tactile feedback to adjust their motion to various objects, but capturing and reproducing the applied forces is an open research question. In this paper we introduce a wearable fingertip tactile sensor, which captures the distributed 3-axis force vectors on the fingertip. The fingertip tactile sensor is interchangeable between the human hand and the robot hand, meaning that it can also be assembled to fit on a robot hand such as the Allegro hand. This paper presents the structural aspects of the sensor as well as the methodology and approach used to design, manufacture, and calibrate the sensor. The sensor is able to measure forces accurately with a mean absolute error of 0.21, 0.16, and 0.44 Newtons in X, Y, and Z directions, respectively.

To support humans in their daily lives, robots are required to autonomously learn, adapt to objects and environments, and perform the appropriate actions. We tackled on the task of cooking scrambled eggs using real ingredients, in which the robot needs to perceive the states of the egg and adjust stirring movement in real time, while the egg is heated and the state changes continuously. In previous works, handling changing objects was found to be challenging because sensory information includes dynamical, both important or noisy information, and the modality which should be focused on changes every time, making it difficult to realize both perception and motion generation in real time. We propose a predictive recurrent neural network with an attention mechanism that can weigh the sensor input, distinguishing how important and reliable each modality is, that realize quick and efficient perception and motion generation. The model is trained with learning from the demonstration, and allows the robot to acquire human-like skills. We validated the proposed technique using the robot, Dry-AIREC, and with our learning model, it could perform cooking eggs with unknown ingredients. The robot could change the method of stirring and direction depending on the status of the egg, as in the beginning it stirs in the whole pot, then subsequently, after the egg started being heated, it starts flipping and splitting motion targeting specific areas, although we did not explicitly indicate them.

Purpose: Manual annotations for training deep learning (DL) models in auto-segmentation are time-intensive. This study introduces a hybrid representation-enhanced sampling strategy that integrates both density and diversity criteria within an uncertainty-based Bayesian active learning (BAL) framework to reduce annotation efforts by selecting the most informative training samples. Methods: The experiments are performed on two lower extremity (LE) datasets of MRI and CT images, focusing on the segmentation of the femur, pelvis, sacrum, quadriceps femoris, hamstrings, adductors, sartorius, and iliopsoas, utilizing a U-net-based BAL framework. Our method selects uncertain samples with high density and diversity for manual revision, optimizing for maximal similarity to unlabeled instances and minimal similarity to existing training data. We assess the accuracy and efficiency using Dice and a proposed metric called reduced annotation cost (RAC), respectively. We further evaluate the impact of various acquisition rules on BAL performance and design an ablation study for effectiveness estimation. Results: In MRI and CT datasets, our method was superior or comparable to existing ones, achieving a 0.8\% Dice and 1.0\% RAC increase in CT (statistically significant), and a 0.8\% Dice and 1.1\% RAC increase in MRI (not statistically significant) in volume-wise acquisition. Our ablation study indicates that combining density and diversity criteria enhances the efficiency of BAL in musculoskeletal segmentation compared to using either criterion alone. Conclusion: Our sampling method is proven efficient in reducing annotation costs in image segmentation tasks. The combination of the proposed method and our BAL framework provides a semi-automatic way for efficient annotation of medical image datasets., Comment: 15 pages, 5 figures

Osteoporosis is a prevalent bone disease that causes fractures in fragile bones, leading to a decline in daily living activities. Dual-energy X-ray absorptiometry (DXA) and quantitative computed tomography (QCT) are highly accurate for diagnosing osteoporosis; however, these modalities require special equipment and scan protocols. To frequently monitor bone health, low-cost, low-dose, and ubiquitously available diagnostic methods are highly anticipated. In this study, we aim to perform bone mineral density (BMD) estimation from a plain X-ray image for opportunistic screening, which is potentially useful for early diagnosis. Existing methods have used multi-stage approaches consisting of extraction of the region of interest and simple regression to estimate BMD, which require a large amount of training data. Therefore, we propose an efficient method that learns decomposition into projections of bone-segmented QCT for BMD estimation under limited datasets. The proposed method achieved high accuracy in BMD estimation, where Pearson correlation coefficients of 0.880 and 0.920 were observed for DXA-measured BMD and QCT-measured BMD estimation tasks, respectively, and the root mean square of the coefficient of variation values were 3.27 to 3.79% for four measurements with different poses. Furthermore, we conducted extensive validation experiments, including multi-pose, uncalibrated-CT, and compression experiments toward actual application in routine clinical practice., Comment: 20 pages and 22 figures