Abstract

We present CheXtriev, a graph-based, anatomy-aware framework for chest radiograph retrieval. Unlike prior methods focussed on global features, our method leverages graph transformers to extract informative features from specific anatomical regions. Furthermore, it captures spatial context and the interplay between anatomical location and findings. This contextualization, grounded in evidence-based anatomy, results in a richer anatomy-aware representation and leads to more accurate, effective and efficient retrieval, particularly for less prevalent findings. CheXtriv outperforms state-of-the-art global and local approaches by to in retrieval accuracy and to in ranking quality.

Abstract

Biological processes like growth, aging, and disease progression are generally studied with follow-up scans taken at different time points, i.e., image time series (TS) based analysis. Image TS represents the evolution of anatomy over time, but different anatomies may have different structural characteristics and temporal paths. Therefore, separating the time-dependent path difference and time-independent basic anatomy/shape changes is important when comparing two image TS to understand the causes of the observed differences better. A method to untangle and quantify the path and shape difference between the TS is presented in this paper. The proposed method is evaluated with simulated and adult and fetal neuro templates. Results show that the metric can separate and quantify the path and shape differences between TS.

Abstract

This paper presents the architecture of a two-wheeler mobile sensing platform aimed at collecting compre-hensive two-wheeler riding data. Two-wheelers, prevalent in many countries for their agility and efficiency lack extensive study in autonomous vehicle research. Recognizing this gap, our work introduces a scalable and modular platform equipped with multiple sensors and high-performance embedded systems. The platform captures essential data points such as GPS, acceleration, gyroscope, speed, and 360-degree image and depth data, crucial for developing deep learning models for autonomous navigation and accident prevention. We detail the hardware architecture, consisting of an Nvidia Jetson TX2 NX platform, Raspberry Pi 4 Model B, and various sensors, all tailored to operate efficiently on a two-wheeler. The software methodology is designed to synchronize and process the data effectively, hence allowing us to generate accurate and thorough datasets. Our results demonstrate the platform's potential in improving two-wheeler safety and contributing to the advancement of autonomous vehicle technologies.

Abstract

No Results Found

Abstract

This work proposes a resource efficient robust control scheme for missile-target engagement scenarios subjected to external disturbances. The robustness is achieved by using an annular event-triggered artificial time-delayed control (ET-TDC) methodology with input saturation. The ET-TDC philosophy uses the TDC strategy through a dynamic predefined triggering mechanism which overcomes the requirement to update the control periodically for every sampling instant, unlike conventional TDC and other robust control schemes. Thus the proposed methodology can tackle uncertainties with minimal a-priori knowledge while significantly reducing the over-utilisation of system resources. In addition, the adopted event-triggered mechanism facilitates further conservation of energy which might be crucial for mid-to-long range engagement scenarios. The closed-loop stability is derived analytically and the simulation results illustrate the efficacy of the proposed guidance framework in comparison with other state-of-art robust control methodologies.

Abstract

Transportation of a suspended payload using quadrotor demands a controller to simultaneously track the desired path and stabilize the planar payload swing angles. Such a control objective is made challenging by the need to orchestrate the coupling between fully actuated dynamics (quadrotor attitude) and underactuated dynamics (quadrotor position and planar payload swings). State-of-the-art controllers cannot orchestrate such coupled dynamics in the presence of unmodeled and state-dependent terms, such as aerodynamic drags and rotor downwash. This article solves this control challenge by adaptively estimating the state-dependent uncertainty. Closed-loop stability for the coupled underactuated and fully actuated dynamics is established analytically. Extensive real-time experiments confirm significant improvements over the state-of-the-art under various scenarios, such as path tracking with suspended payload and anti-swing control against externally induced payload swings.

Abstract

Despite participants engaging in single modality stimuli, such as watching images or silent videos, recent work has demonstrated that multi-modal Transformer models can predict visual brain activity impressively well, even with incongruent modality representations. This raises the question of how accurately these multi-modal models can predict brain activity when participants are engaged in multi-modal stimuli. As these models grow increasingly popular, their use in studying neural activity provides insights into how our brains respond to such multi-modal naturalistic stimuli, i.e., where it separates and integrates information from different sensory modalities. We investigate this question by using multiple unimodal and two types of multi-modal models—cross-modal and jointly pretrained—to determine which type of models is more relevant to fMRI brain activity when participants were engaged in watching movies (videos with audio). We observe that both types of multi-modal models show improved alignment in several language and visual regions. This study also helps in identifying which brain regions process unimodal versus multi-modal information. We further investigate the impact of removal of unimodal features from multi-modal representations and find that there is additional information beyond the unimodal embeddings that is processed in the visual and language regions. Based on this investigation, we find that while for cross-modal models, their brain alignment is partially attributed to the video modality; for jointly pretrained models, it is partially attributed to both the video and audio modalities.

Abstract

LOCALITY and INTERFERENCE are two mechanisms which are attested to drive sentence comprehension. However, the relationship between them remains unclear—are they alternative explanations or do they operate independently? To answer this question, we test the hypothesis that in Hindi, interference effects (measured by semantic similarity and case markers) significantly predict locality effects (modelled using dependency length quantifying distance between syntactic heads and their dependents) within a sentence, while controlling for expectation-based measures and discourse givenness. Using data from the HindiUrdu Treebank corpus (HUTB), we validate the stated hypothesis. We demonstrate that sentences with longer dependency length consistently have semantically similar preverbal dependents, more case markers, greater syntactic surprisal, and violate intra-sentential givenness considerations. Overall, our findings point towards the conclusion that locality effects are reducible to broader memory interference effects rather than being distinct manifestations of locality in syntax. Finally, we discuss the implications of our findings for the theories of interference in comprehension.

Abstract

Methods and systems for obtaining end-to-end classification model using mixture density network

Abstract

Deep learning models are capable of complex auditory processing tasks such as keyword spotting, genre classification, and audio captioning, yet remain opaque. While several works have explored interpretability of neural networks for computer vision and natural language processing, the audio modality has been largely ignored. In this paper, we study the behavior of the audio CNN encoder used in the contrastively trained language-audio model, CLAP. In the domain of music and human speech sounds, we localize and identify the layers of the network that perform well on tasks of varying complexity, sometimes even outperforming the model's final outputs. Digging deeper, we also localize specific dataset classes to neuron clusters within a layer and analyze a cluster’s contribution to the model’s discriminability for that class. To perform these analyses, we propose an automated framework that can leverage a small dataset of a few thousand samples to evaluate and score neuron clusters for their role in classification. Our findings provide insights into the hierarchical nature of representations in audio CNNs, paving the way for improved interpretability of audio model.