Objective: The objective of this work is an attempt towards non-contact freehand 3D ultrasound imaging with minimal complexity added to the existing point of care ultrasound (POCUS) systems. Methods: This study proposes a novel approach of using a mechanical track for non-contact ultrasound (US) scanning. The approach thus restricts the probe motion to a linear plane, to simplify the acquisition and 3D reconstruction process. A pipeline for US 3D volume reconstruction employing an US research platform and a GPU-based edge device is developed. Results: The efficacy of the proposed approach is demonstrated through ex-vivo and in-vivo experiments. Conclusion: The proposed approach with the adjustable field of view capability, non-contact design, and low cost of deployment without significantly altering the existing setup would open doors for up gradation of traditional systems to a wide range of 3D US imaging applications. Significance: Ultrasound (US) imaging is a popular clinical imaging modality for the point-of-care bedside imaging, particularly of the wrist/knee in the pediatric population due to its non-invasive and radiation free nature. However, the limited views of tissue structures obtained with 2D US in such scenarios make the diagnosis challenging. To overcome this, 3D US imaging which uses 2D US images and their orientation/position to reconstruct 3D volumes was developed. The accurate position estimation of the US probe at low cost has always stood as a challenging task in 3D reconstruction. Additionally, US imaging involves contact, which causes difficulty to pediatric subjects while monitoring live fractures or open wounds. Towards overcoming these challenges, a novel framework is attempted in this work.
This work investigates the predictive potential of bipolar electroencephalogram (EEG) recordings towards efficient prediction of poor neurological outcomes. A retrospective design using a hybrid deep learning approach is utilized to optimize an objective function aiming for high specificity, i.e., true positive rate (TPR) with reduced false positives (< 0.05). A multi-channel EEG array of 18 bipolar channel pairs from a randomly selected 5-minute segment in an hour is kept. In order to determine the outcome prediction, a combination of a feature encoder with 1-D convolutional layers, learnable position encoding, a context network with attention mechanisms, and finally, a regressor and classifier blocks are used. The feature encoder extricates local temporal and spatial features, while the following position encoding and attention mechanisms attempt to capture global temporal dependencies. Results: The proposed framework by our team, OUS IVS, when validated on the challenge hidden validation data, exhibited a score of 0.57.
Preventing the performance decay of Transformers on inputs longer than those used for training has been an important challenge in extending the context length of these models. Though the Transformer architecture has fundamentally no limits on the input sequence lengths it can process, the choice of position encoding used during training can limit the performance of these models on longer inputs. We propose a novel functional relative position encoding with progressive interpolation, FIRE, to improve Transformer generalization to longer contexts. We theoretically prove that this can represent some of the popular relative position encodings, such as T5's RPE, Alibi, and Kerple. We next empirically show that FIRE models have better generalization to longer contexts on both zero-shot language modeling and long text benchmarks.
We present a factor graph formulation and particle-based sum-product algorithm for robust localization and tracking in multipath-prone environments. The proposed sequential algorithm jointly estimates the mobile agent's position together with a time-varying number of multipath components (MPCs). The MPCs are represented by "delay biases" corresponding to the offset between line-of-sight (LOS) component delay and the respective delays of all detectable MPCs. The delay biases of the MPCs capture the geometric features of the propagation environment with respect to the mobile agent. Therefore, they can provide position-related information contained in the MPCs without explicitly building a map of the environment. We demonstrate that the position-related information enables the algorithm to provide high-accuracy position estimates even in fully obstructed line-of-sight (OLOS) situations. Using simulated and real measurements in different scenarios we demonstrate that the proposed algorithm significantly outperforms state-of-the-art multipath-aided tracking algorithms and show that the performance of our algorithm constantly attains the posterior Cramer-Rao lower bound (P-CRLB). Furthermore, we demonstrate the implicit capability of the proposed method to identify unreliable measurements and, thus, to mitigate lost tracks.
Non-ideal position estimation results in degraded performance of synchronous motor drive systems due to reduction of the average capability of the drive as well as torque harmonics of different orders. The signature and extent of the performance degradation is further dependent, quite significantly, on the current control architecture, i.e., feedforward or feedback control, employed. This paper presents a comprehensive analysis of non-idealities or errors in position estimation and their effects on the control performance of synchronous motor drives. Analytical models capturing the error in various signals caused by position sensing errors in the drive system for different control architectures are presented and are validated with simulation and experimental results on a prototype permanent magnet synchronous motor drive.
Field oriented control of permanent magnet synchronous motor drives involves the closed-loop regulation of currents in the synchronous reference frame. The current feedback is directly affected by errors in both position and stationary frame current measurements. This paper presents the exact analytical expression for estimated synchronous frame currents under simultaneous errors in both sensors along with a detailed analysis of the incorrect estimation on the closed-loop current control performance.
There is a niche of companies responsible for intermediating the purchase of large batches of varied products for other companies, for which the main challenge is to perform product description standardization, i.e., matching an item described by a client with a product described in a catalog. The problem is complex since the client's product description may be: (1) potentially noisy; (2) short and uninformative (e.g., missing information about model and size); and (3) cross-language. In this paper, we formalize this problem as a ranking task: given an initial client product specification (query), return the most appropriate standardized descriptions (response). In this paper, we propose TPDR, a two-step Transformer-based Product and Class Description Retrieval method that is able to explore the semantic correspondence between IS and SD, by exploiting attention mechanisms and contrastive learning. First, TPDR employs the transformers as two encoders sharing the embedding vector space: one for encoding the IS and another for the SD, in which corresponding pairs (IS, SD) must be close in the vector space. Closeness is further enforced by a contrastive learning mechanism leveraging a specialized loss function. TPDR also exploits a (second) re-ranking step based on syntactic features that are very important for the exact matching (model, dimension) of certain products that may have been neglected by the transformers. To evaluate our proposal, we consider 11 datasets from a real company, covering different application contexts. Our solution was able to retrieve the correct standardized product before the 5th ranking position in 71% of the cases and its correct category in the first position in 80% of the situations. Moreover, the effectiveness gains over purely syntactic or semantic baselines reach up to 3.7 times, solving cases that none of the approaches in isolation can do by themselves.
With their unique combination of characteristics - an energy density almost 100 times that of human muscle, and a power density of 5.3 kW/kg, similar to a jet engine's output - Nylon artificial muscles stand out as particularly apt for robotics applications. However, the necessity of integrating sensors and controllers poses a limitation to their practical usage. Here we report a constant power open-loop controller based on machine learning. We show that we can control the position of a nylon artificial muscle without external sensors. To this end, we construct a mapping from a desired displacement trajectory to a required power using an ensemble encoder-style feed-forward neural network. The neural controller is carefully trained on a physics-based denoised dataset and can be fine-tuned to accommodate various types of thermal artificial muscles, irrespective of the presence or absence of hysteresis.
Based on the theory of homogeneous spaces we derive geometrically optimal edge attributes to be used within the flexible message-passing framework. We formalize the notion of weight sharing in convolutional networks as the sharing of message functions over point-pairs that should be treated equally. We define equivalence classes of point-pairs that are identical up to a transformation in the group and derive attributes that uniquely identify these classes. Weight sharing is then obtained by conditioning message functions on these attributes. As an application of the theory, we develop an efficient equivariant group convolutional network for processing 3D point clouds. The theory of homogeneous spaces tells us how to do group convolutions with feature maps over the homogeneous space of positions $\mathbb{R}^3$, position and orientations $\mathbb{R}^3 {\times} S^2$, and the group $SE(3)$ itself. Among these, $\mathbb{R}^3 {\times} S^2$ is an optimal choice due to the ability to represent directional information, which $\mathbb{R}^3$ methods cannot, and it significantly enhances computational efficiency compared to indexing features on the full $SE(3)$ group. We support this claim with state-of-the-art results -- in accuracy and speed -- on five different benchmarks in 2D and 3D, including interatomic potential energy prediction, trajectory forecasting in N-body systems, and generating molecules via equivariant diffusion models.
Current language models often fail to incorporate long contexts efficiently during generation. We show that a major contributor to this issue are attention priors that are likely learned during pre-training: relevant information located earlier in context is attended to less on average. Yet even when models fail to use the information from a relevant document in their response, they still pay preferential attention to that document compared to an irrelevant document at the same position. We leverage this fact to introduce ``attention sorting'': perform one step of decoding, sort documents by the attention they receive (highest attention going last), repeat the process, generate the answer with the newly sorted context. We find that attention sorting improves performance of long context models. Our findings highlight some challenges in using off-the-shelf language models for retrieval augmented generation.
Whole slide image (WSI) classification is a critical task in computational pathology. However, the gigapixel-size of such images remains a major challenge for the current state of deep-learning. Current methods rely on multiple-instance learning (MIL) models with frozen feature extractors. Given the the high number of instances in each image, MIL methods have long assumed independence and permutation-invariance of patches, disregarding the tissue structure and correlation between patches. Recent works started studying this correlation between instances but the computational workload of such a high number of tokens remained a limiting factor. In particular, relative position of patches remains unaddressed. We propose to apply a straightforward encoding module, namely a RoFormer layer , relying on memory-efficient exact self-attention and relative positional encoding. This module can perform full self-attention with relative position encoding on patches of large and arbitrary shaped WSIs, solving the need for correlation between instances and spatial modeling of tissues. We demonstrate that our method outperforms state-of-the-art MIL models on three commonly used public datasets (TCGA-NSCLC, BRACS and Camelyon16)) on weakly supervised classification tasks. Code is available at https://github.com/Sanofi-Public/DDS-RoFormerMIL
Large language models (LLMs) have shown promise as automated evaluators for assessing the quality of answers generated by AI systems. However, these LLM-based evaluators exhibit position bias, or inconsistency, when used to evaluate candidate answers in pairwise comparisons, favoring either the first or second answer regardless of content. To address this limitation, we propose PORTIA, an alignment-based system designed to mimic human comparison strategies to calibrate position bias in a lightweight yet effective manner. Specifically, PORTIA splits the answers into multiple segments, aligns similar content across candidate answers, and then merges them back into a single prompt for evaluation by LLMs. We conducted extensive experiments with six diverse LLMs to evaluate 11,520 answer pairs. Our results show that PORTIA markedly enhances the consistency rates for all the models and comparison forms tested, achieving an average relative improvement of 47.46%. Remarkably, PORTIA enables less advanced GPT models to achieve 88% agreement with the state-of-the-art GPT-4 model at just 10% of the cost. Furthermore, it rectifies around 80% of the position bias instances within the GPT-4 model, elevating its consistency rate up to 98%. Subsequent human evaluations indicate that the PORTIA-enhanced GPT-3.5 model can even surpass the standalone GPT-4 in terms of alignment with human evaluators. These findings highlight PORTIA's ability to correct position bias, improve LLM consistency, and boost performance while keeping cost-efficiency. This represents a valuable step toward a more reliable and scalable use of LLMs for automated evaluations across diverse applications.
Switched Reluctance Motors (SRMs) enable power-efficient actuation with mechanically simple designs. This paper aims to identify the nonlinear relationship between torque, rotor angle, and currents, to design commutation functions that minimize torque ripple in SRMs. This is achieved by conducting specific closed-loop experiments using purposely imperfect commutation functions and identifying the nonlinear dynamics via Bayesian estimation. A simulation example shows that the presented method is robust to position-dependent disturbances, and experiments suggest that the identification method enables the design of commutation functions that significantly increase performance. The developed approach enables accurate identification of the torque-current-angle relationship in SRMs, without the need for torque sensors, an accurate linear model, or an accurate model of position-dependent disturbances, making it easy to implement in production.
The motion visualization in a structural component was studied for defect detection. Elastic motions were excited by hammer impacts at multiple points and received by an accelerometer at a fixed point. Reciprocity in elastodynamics is only valid under certain conditions. Its validity under given experimental conditions was derived from the elastodynamic reciprocity theorem. Based on this, the dynamic motion of the structural component was obtained for fixed-point excitation from measurements performed using multipoint excitations. In the visualized eigenmodes, significant additional deformation was observed at the wall thinning inserted as an artificial defect. To prevent the dependence of defect detection on its position within the mode shape, another approach was proposed based on the extraction of guided wave modes immediately after impact excitation. It is shown that this maximum intensity projection method works well in detecting defects.
In this work, we propose a position and orientation-aware one-shot learning framework for medical action recognition from signal data. The proposed framework comprises two stages and each stage includes signal-level image generation (SIG), cross-attention (CsA), dynamic time warping (DTW) modules and the information fusion between the proposed privacy-preserved position and orientation features. The proposed SIG method aims to transform the raw skeleton data into privacy-preserved features for training. The CsA module is developed to guide the network in reducing medical action recognition bias and more focusing on important human body parts for each specific action, aimed at addressing similar medical action related issues. Moreover, the DTW module is employed to minimize temporal mismatching between instances and further improve model performance. Furthermore, the proposed privacy-preserved orientation-level features are utilized to assist the position-level features in both of the two stages for enhancing medical action recognition performance. Extensive experimental results on the widely-used and well-known NTU RGB+D 60, NTU RGB+D 120, and PKU-MMD datasets all demonstrate the effectiveness of the proposed method, which outperforms the other state-of-the-art methods with general dataset partitioning by 2.7%, 6.2% and 4.1%, respectively.
In treating gastrointestinal cancer using radiotherapy, the role of the radiation oncologist is to administer high doses of radiation, through x-ray beams, toward the tumor while avoiding the stomach and intestines. With the advent of precise radiation treatment technology such as the MR-Linac, oncologists can visualize the daily positions of the tumors and intestines, which may vary day to day. Before delivering radiation, radio oncologists must manually outline the position of the gastrointestinal organs in order to determine position and direction of the x-ray beam. This is a time consuming and labor intensive process that may substantially prolong a patient's treatment. A deep learning (DL) method can automate and expedite the process. However, many deep neural networks approaches currently in use are black-boxes which lack interpretability which render them untrustworthy and impractical in a healthcare setting. To address this, an emergent field of AI known as Explainable AI (XAI) may be incorporated to improve the transparency and viability of a model. This paper proposes a deep learning pipeline that incorporates XAI to address the challenges of organ segmentation.
AI is becoming increasingly popular in artistic practices, but the tools for informing practitioners about the environmental impact (and other sustainability implications) of AI are adapted for other contexts than creative practices -- making the tools and sustainability implications of AI not accessible for artists and creative practitioners. In this position paper, I describe two empirical studies that aim to develop environmental sustainability reflection systems for AI Arts, and discuss and introduce Explainable Sustainability in for AI Arts.
Weight-sharing quantization has emerged as a technique to reduce energy expenditure during inference in large neural networks by constraining their weights to a limited set of values. However, existing methods for weight-sharing quantization often make assumptions about the treatment of weights based on value alone that neglect the unique role weight position plays. This paper proposes a probabilistic framework based on Bayesian neural networks (BNNs) and a variational relaxation to identify which weights can be moved to which cluster centre and to what degree based on their individual position-specific learned uncertainty distributions. We introduce a new initialisation setting and a regularisation term which allow for the training of BNNs under complex dataset-model combinations. By leveraging the flexibility of weight values captured through a probability distribution, we enhance noise resilience and downstream compressibility. Our iterative clustering procedure demonstrates superior compressibility and higher accuracy compared to state-of-the-art methods on both ResNet models and the more complex transformer-based architectures. In particular, our method outperforms the state-of-the-art quantization method top-1 accuracy by 1.6% on ImageNet using DeiT-Tiny, with its 5 million+ weights now represented by only 296 unique values.
A novel over-the-air computation (AirComp) framework, empowered by the incorporation of movable antennas (MAs), is proposed to significantly enhance computation accuracy. Within this framework, the joint optimization of transmit power control, antenna positioning, and receive combining is investigated. An efficient method is proposed to tackle the problem of computation mean-squared error (MSE) minimization, capitalizing on the approach of alternating optimization. Numerical results are provided to substantiate the superior MSE performance of the proposed framework, which establish its clear advantage over benchmark systems employing conventional fixed-position antennas (FPAs).
Self-attention-based vision transformers (ViTs) have emerged as a highly competitive architecture in computer vision. Unlike convolutional neural networks (CNNs), ViTs are capable of global information sharing. With the development of various structures of ViTs, ViTs are increasingly advantageous for many vision tasks. However, the quadratic complexity of self-attention renders ViTs computationally intensive, and their lack of inductive biases of locality and translation equivariance demands larger model sizes compared to CNNs to effectively learn visual features. In this paper, we propose a light-weight and efficient vision transformer model called DualToken-ViT that leverages the advantages of CNNs and ViTs. DualToken-ViT effectively fuses the token with local information obtained by convolution-based structure and the token with global information obtained by self-attention-based structure to achieve an efficient attention structure. In addition, we use position-aware global tokens throughout all stages to enrich the global information, which further strengthening the effect of DualToken-ViT. Position-aware global tokens also contain the position information of the image, which makes our model better for vision tasks. We conducted extensive experiments on image classification, object detection and semantic segmentation tasks to demonstrate the effectiveness of DualToken-ViT. On the ImageNet-1K dataset, our models of different scales achieve accuracies of 75.4% and 79.4% with only 0.5G and 1.0G FLOPs, respectively, and our model with 1.0G FLOPs outperforms LightViT-T using global tokens by 0.7%.