Hahn 477 High Performance Wearable Ultrasound as a Human-Machine Interface for Wrist and Hand Kinematic Tracking .

T-BME publishes basic and applied papers dealing with biomedical engineering. Papers range from engineering development in methods and techniques with biomedical applications to experimental and clinical investigations with engineering contributions.

Accurate visual classification of bladder tissue during Trans-Urethral Resection of Bladder Tumor (TURBT) procedures is essential to improve early cancer diagnosis and treatment. During TURBT interventions, White Light Imaging (WLI) and Narrow Band Imaging (NBI) techniques are used for lesion detection. Each imaging technique provides diverse visual information that allows clinicians to identify and classify cancerous lesions. Computer vision methods that use both imaging techniques could improve endoscopic diagnosis. We address the challenge of tissue classification when annotations are available only in one domain, in our case WLI, and the endoscopic images correspond to an unpaired dataset, i.e. there is no exact equivalent for every image in both NBI and WLI domains. We propose a semi-surprised Generative Adversarial Network (GAN)-based method composed of three main components: a teacher network trained on the labeled WLI data; a cycle-consistency GAN to perform unpaired image-to-image translation, and a multi-input student network. To ensure the quality of the synthetic images generated by the proposed GAN we perform a detailed quantitative, and qualitative analysis with the help of specialists. The overall average classification accuracy, precision, and recall obtained with the proposed method for tissue classification are 0.90, 0.88, and 0.89 respectively, while the same metrics obtained in the unlabeled domain (NBI) are 0.92, 0.64, and 0.94 respectively. The quality of the generated images is reliable enough to deceive specialists. This study shows the potential of using semi-supervised GAN-based bladder tissue classification when annotations are limited in multi-domain data.

One-shot organ segmentation (OS2) aims at segmenting the desired organ regions from the input medi-cal imaging data with only one pre-annotated example as the reference. By using the minimal annotation data to facilitate organ segmentation, OS2 receives great attention in the medical image analysis community due to its weak re-quirement on human annotation. In OS2, one core issue is to explore the mutual information between the support (ref-erence slice) and the query (test slice). Existing methods rely heavily on the similarity between slices, and additional slice allocation mechanisms need to be designed to reduce the impact of the similarity between slices on the segmen-tation performance. To address this issue, we build a novel support-query interactive embedding (SQIE) module, which is equipped with the channel-wise co-attention, spatial-wise co-attention, and spatial bias transformation blocks to identify "what to look", "where to look", and "how to look" in the input test slice. By combining the three mechanisms, we can mine the interactive information of the intersection area and the disputed area between slices, and establish the feature connection between the target in slices with low similarity. We also propose a self-supervised contrastive learning framework, which transforms knowledge from the physical position to the embedding space to facilitate the self-supervised interactive embedding of the query and support slices. Comprehensive experiments on two large benchmarks demonstrate the superior capacity of the pro-posed approach when compared with the current alterna-tives and baseline models.

Musculoskeletal health monitoring is limited in everyday settings where patient symptoms can substantially change - delaying treatment and worsening patient outcomes. Wearable technologies aim to quantify musculoskeletal health outside clinical settings but sensor constraints limit usability. Wearable localized multi-frequency bioimpedance assessment (MFBIA) shows promise for tracking musculoskeletal health but relies on gel electrodes, hindering extended at-home use. Here, we address this need for usable technologies for at-home musculoskeletal health assessment by designing a wearable adhesive-free MFBIA system using textile electrodes in extended uncontrolled mid-activity settings. An adhesive-free multimodal wearable leg MFBIA system was developed in-lab under realistic conditions (5 participants, 45 measurements). Mid-activity textile and gel electrode MFBIA was compared across multiple compound movements (10 participants). Accuracy in tracking long-term changes in leg MFBIA was assessed by correlating gel and textile MFBIA simultaneously recorded in uncontrolled settings (10 participants, 80+ measurement hours). Mid-activity MFBIA measurements with textile electrodes agreed highly with (ground truth) gel electrode measurements (average [Formula: see text], featuring <1-Ohm differences (0.618 ± 0.340 Ω) across all movements. Longitudinal MFBIA changes were successfully measured in extended at-home settings (repeated measures r = 0.84). Participant responses found the system to be comfortable and intuitive (8.3/10), and all participants were able to don and operate the system independently. This work demonstrates wearable textile electrodes can be a viable substitute for gel electrodes when monitoring leg MFBIA in dynamic, uncontrolled settings. Adhesive-free MFBIA can improve healthcare by enabling robust wearable musculoskeletal health monitoring in at-home and everyday settings.

Surface electromyography (sEMG) patterns have been decoded using learning-based methods that determine complicated nonlinear decision boundaries. However, overlapping classes in sEMG pattern recognition still degrade the classification accuracy because they cannot be separated by the decision boundaries. We hypothesized that certain overlapping classes can be separated while tracing the temporal history of sEMG patterns. Therefore, a novel post-processing method is proposed to adjust classification errors using the separated patterns from the temporal history of overlapping classes. The proposed method confirms the confidence of the prediction result by calculating the instantaneous pattern separability for the sequential sEMG input. The prediction result with high separability pattern is considered to have a high confidence of being correct (reliable). This result is stored for adjusting the next sEMG input. When the subsequent prediction is identified as having low confidence (unreliable), the predicted result is adjusted using the stored reliable predicted results. The proposed method adds an adjustment step to an existing classifier (maximum likelihood classifier (MLC), k-nearest neighbor (KNN), and support vector machine (SVM)), such that it can be attached to the back-end regardless of the type of classifier. Ten subjects performed 13 types of hand gestures, including overlapping patterns. The overall classification accuracy was enhanced to 88.93%(+8.12%p, MLC), 91.31%(+7.68%p, KNN), and 99.65%(+11.63%p, SVM) after the implementation of the proposed post-processing. Additionally, a faster and more accurate gesture classification was achieved with accuracy enhancement before gesture completion as 85.62%(+4.23%p, MLC), 89.77%(+4.23%p, KNN), and 97.62%(+11.12%p, SVM).

Heart failure, respiratory failure and kidney failure are three severe organ failures (OF) that have high mortalities and are most prevalent in intensive care units. The objective of this work is to offer insights into OF clustering from the aspects of graph neural networks and diagnosis history. This paper proposes a neural network-based pipeline to cluster three types of organ failure patients by incorporating embedding pre-train using an ontology graph of the International Classification of Diseases (ICD) codes. We employ an autoencoder-based deep clustering architecture jointly trained with a K-means loss, and a non-linear dimension reduction is performed to obtain patient clusters on the MIMIC-III dataset. The clustering pipeline shows superior performance on a public-domain image dataset. On the MIMIC-III dataset, it discovers two distinct clusters that exhibit different comorbidity spectra which can be related to the severity of diseases. The proposed pipeline is compared with several other clustering models and shows superiority. Our proposed pipeline gives stable clusters, however, they do not correspond to the type of OF which indicates these OF share significant hidden characteristics in diagnosis. These clusters can be used to signal possible complications and severity of illness and aid personalised treatment. We are the first to apply an unsupervised approach to offer insights from a biomedical engineering perspective on these three types of organ failure, and publish the pre-trained embeddings for future transfer learning.

Here, we demonstrate a technique to control afterload while monitoring contractile force exerted by EHTs. We developed an apparatus that can regulate EHT boundary conditions using real-time feedback control. The system is comprised of a pair of piezoelectric actuators that can strain the scaffold and a microscope that can measure EHT force and length. Closed loop control allows dynamic regulation of effective EHT boundary stiffness. When controlled to switch instantaneously from auxotonic to isometric boundary conditions, EHT twitch force immediately doubled. Changes in EHT twitch force as a function of effective boundary stiffness were characterized and compared to twitch force in auxotonic conditions. EHT contractility can be regulated dynamically through feedback control of effective boundary stiffness. The capacity to alter the mechanical boundary conditions of an engineered tissue dynamically offers a new way to probe tissue mechanics. This could be used to mimic afterload changes that occur naturally in disease, or to improve mechanical techniques for EHT maturation.

This study aimed to propose a new clinical modality for the relief of in-stent restenosis (ISR) using focused ultrasound (FUS) ablation. In the first research stage, a miniaturized FUS device was developed for the sonification of the remaining plaque after stenting, known as one of the causes of ISR. This study presents a miniaturized (<2.8 mm) intravascular FUS transducer for ISR treatment. The performance of the transducer was predicted through a structural-acoustic simulation, followed by fabrication of the prototype device. Using the prototype FUS transducer, we demonstrated tissue ablation with bio-tissues over metallic stents, mimicking in-stent tissue ablation. Next, we conducted a safety test by detecting the existence of thermal damage to the arterial tissue upon sonication with a controlled dose. The prototype device successfully delivered sufficient acoustic intensity (> 30 W/cm2) to a bio tissue (chicken breast) through a metallic stent. The ablation volume was approximately 3.9×7.8×2.6 mm3. Furthermore, 1.5 min sonication was sufficient to obtain an ablating depth of approximately 1.0 mm, not thermally damaging the underlying artery vessel. Conclusion: We demonstrated in-stent tissue sonoablation, suggesting it could be as a future ISR treatment modality. Significance: Comprehensive test results provide a key understanding of FUS applications using metallic stents. Furthermore, the developed device can be used for sonoablation of the remaining plaque, providing a novel approach to the treatment of ISR.