arXiv subjects

Satrajit S. Ghosh

Publications and source records attributed to Satrajit S. Ghosh.

7 recordsLinked to original sources

arXiv2026

An Intelligent Infrastructure as a Foundation for Modern Science

Infrastructure shapes societies and scientific discovery. Traditional scientific infrastructure, often static and fragmented, leads to issues like data silos, lack of interoperability and reproducibility, and unsustainable short-lived solutions. Our current technical inability and social reticence to connect and coordinate scientific research and engineering lead to inefficiencies and impede progress. With AI technologies changing how we interact with the world around us, there is an opportunity to transform scientific processes. Neuroscience's exponential growth of multimodal and multiscale data, together with its urgent clinical relevance, demands an adaptive infrastructure that can expose computable states, coordinate across systems, and improve through use. Using neuroscience as a stress test, this perspective argues for a paradigm shift: infrastructure must evolve into a dynamic, AI-aligned ecosystem to accelerate science. Building on several existing principles for data, collective benefit, and digital repositories, I recommend operational guidelines for implementing these principles to create this dynamic ecosystem, aiming to foster a decentralized, self-learning, and self-correcting system where humans and AI can collaborate seamlessly. Addressing the chronic underfunding of scientific infrastructure, acknowledging diverse contributions beyond publications, and coordinating global efforts are critical for this transformation. A coordinating role, even more than analysis, is where AI becomes transformative rather than merely assistive. By prioritizing an intelligent infrastructure as a central scientific instrument for knowledge generation, we can overcome current limitations, accelerate discovery, ensure reproducibility and ethical practices, and ultimately translate neuroscientific understanding into tangible societal benefits, setting a blueprint for other scientific domains.

q-bio.NC↗

arXiv2026

STRUCTSENSE: A Task-Agnostic Agentic Framework for Structured Information Extraction with Human-In-The-Loop Evaluation and Benchmarking

Extracting structured information from scientific literature is critical for accelerating discovery, yet Large Language Models (LLMs) often struggle in specialized domains that require expert knowledge and generalize poorly across tasks. We introduce \textsc{StructSense}, a modular, task-agnostic, open-source framework that integrates ontology-guided symbolic knowledge, agentic self-evaluative refinement, and human-in-the-loop validation for robust domain-aware extraction. We evaluate \textsc{StructSense} on three tasks of increasing semantic complexity: schema-based extraction of assessment instruments (91--100\% accuracy), metadata and resource extraction from scientific papers (86--93\% overall), and named entity recognition (NER) from neuroscience literature (58--75\% label accuracy across 8,882 entities). On two biomedical NER benchmarks (NCBI Disease and S800 Species), the system achieves $\geq$90\% relaxed recall and 62.5--85.8\% strict recall while extracting 1,000--3,600 additional entities beyond gold annotations. The local concept mapping service achieves Hits@1 of 62--82\% under strict matching and 68--86\% under semantic matching. These results across three domains demonstrate that \textsc{StructSense} generalizes across tasks while maintaining source grounding and provenance transparency.

cs.CL↗

arXiv2024

Explaining Deep Learning Embeddings for Speech Emotion Recognition by Predicting Interpretable Acoustic Features

Pre-trained deep learning embeddings have consistently shown superior performance over handcrafted acoustic features in speech emotion recognition (SER). However, unlike acoustic features with clear physical meaning, these embeddings lack clear interpretability. Explaining these embeddings is crucial for building trust in healthcare and security applications and advancing the scientific understanding of the acoustic information that is encoded in them. This paper proposes a modified probing approach to explain deep learning embeddings in the SER space. We predict interpretable acoustic features (e.g., f0, loudness) from (i) the complete set of embeddings and (ii) a subset of the embedding dimensions identified as most important for predicting each emotion. If the subset of the most important dimensions better predicts a given emotion than all dimensions and also predicts specific acoustic features more accurately, we infer those acoustic features are important for the embedding model for the given task. We conducted experiments using the WavLM embeddings and eGeMAPS acoustic features as audio representations, applying our method to the RAVDESS and SAVEE emotional speech datasets. Based on this evaluation, we demonstrate that Energy, Frequency, Spectral, and Temporal categories of acoustic features provide diminishing information to SER in that order, demonstrating the utility of the probing classifier method to relate embeddings to interpretable acoustic features.

cs.SD↗

arXiv2024

The Past, Present, and Future of the Brain Imaging Data Structure (BIDS)

The Brain Imaging Data Structure (BIDS) is a community-driven standard for the organization of data and metadata from a growing range of neuroscience modalities. This paper is meant as a history of how the standard has developed and grown over time. We outline the principles behind the project, the mechanisms by which it has been extended, and some of the challenges being addressed as it evolves. We also discuss the lessons learned through the project, with the aim of enabling researchers in other domains to learn from the success of BIDS.

q-bio.OT↗

arXiv2019

Knowing what you know in brain segmentation using Bayesian deep neural networks

In this paper, we describe a Bayesian deep neural network (DNN) for predicting FreeSurfer segmentations of structural MRI volumes, in minutes rather than hours. The network was trained and evaluated on a large dataset (n = 11,480), obtained by combining data from more than a hundred different sites, and also evaluated on another completely held-out dataset (n = 418). The network was trained using a novel spike-and-slab dropout-based variational inference approach. We show that, on these datasets, the proposed Bayesian DNN outperforms previously proposed methods, in terms of the similarity between the segmentation predictions and the FreeSurfer labels, and the usefulness of the estimate uncertainty of these predictions. In particular, we demonstrated that the prediction uncertainty of this network at each voxel is a good indicator of whether the network has made an error and that the uncertainty across the whole brain can predict the manual quality control ratings of a scan. The proposed Bayesian DNN method should be applicable to any new network architecture for addressing the segmentation problem.

cs.CV↗

arXiv2019

Distributed Weight Consolidation: A Brain Segmentation Case Study

Collecting the large datasets needed to train deep neural networks can be very difficult, particularly for the many applications for which sharing and pooling data is complicated by practical, ethical, or legal concerns. However, it may be the case that derivative datasets or predictive models developed within individual sites can be shared and combined with fewer restrictions. Training on distributed data and combining the resulting networks is often viewed as continual learning, but these methods require networks to be trained sequentially. In this paper, we introduce distributed weight consolidation (DWC), a continual learning method to consolidate the weights of separate neural networks, each trained on an independent dataset. We evaluated DWC with a brain segmentation case study, where we consolidated dilated convolutional neural networks trained on independent structural magnetic resonance imaging (sMRI) datasets from different sites. We found that DWC led to increased performance on test sets from the different sites, while maintaining generalization performance for a very large and completely independent multi-site dataset, compared to an ensemble baseline.

cs.LG↗

arXiv2012

Algorithm to suppress scanner noise in recorded speech during functional magnetic resonance imaging

The high-intensity, repetitive noise associated with functional magnetic resonance imaging hinders on-line monitoring of subjects' speech and/or recording speech signals suitable for off-line analysis. The proposed algorithm enhances the speech signal by suppressing the scanner noise in the signal recorded by a single-channel microphone. Significant increases in signal-to-noise ratio are achieved using an adaptive filter that combines time and frequency domain elements. In addition to providing a recording suitable for speech analysis, such a real-time system provides an alternative means (to, e.g., the "panic ball") for communication between the patient and the operator during image acquisition.

cs.SD↗