arXiv subjects

Muhammad Irshad

Publications and source records attributed to Muhammad Irshad.

3 recordsLinked to original sources

arXiv2026

MA-RAG: Multi-Agent Retrieval-Augmented Generation for Query-Driven Summarization of Longitudinal Parkinson's Disease Assessments

Accurate interpretation of single-visit and longitudinal clinical assessments for Parkinson's disease is time-consuming and often depends on specialist expertise. Although large language models (LLMs) can generate natural language summaries, they frequently lack domain-specific clinical grounding and struggle to produce factually correct and temporally consistent responses for structured longitudinal assessment data. To address these limitations, we propose MA-RAG, a query-driven multi-agent retrieval-augmented generation framework that decomposes clinical reasoning into domain-specialized agents, combines structured fact extraction, and synthesizes clinically grounded summaries through a final verification stage. The framework supports four clinical analysis tasks: single-session, trajectory, comparison, and cohort summarization. We evaluate MA-RAG using objective metrics, namely Fact Precision, Hallucination Rate, Temporal Fidelity, and Semantic Similarity, together with subjective evaluations conducted by clinical experts. Compared to Traditional, RAG-only, and Single-agent RAG baselines, MA-RAG substantially improves factual correctness, achieving up to a 122% relative increase in Fact Precision (from 0.436 to 0.990) and reducing the Hallucination Rate by up to 98% (from 0.564 to 0.010), while consistently receiving top ratings from clinical experts for organization and clinical usefulness. These results demonstrate that domain-specialized multi-agent reasoning enables reliable query-driven summarization of structured longitudinal clinical assessment data.

cs.CL↗

arXiv2026

You Don't Need Attention: Gated Convolutional Modeling for Watch-Based Fall Detection

Existing deep learning approaches for wearable fall detection systems rely on self-attention mechanisms that impose quadratic computational overhead, distributing weights across all time steps. This global weight distribution impairs the precise localization of the brief impact signatures that characterize falls within short, fixed-length windows. To overcome this challenge, we propose Gated-CNN, a lightweight dual-stream architecture that processes accelerometer and gyroscope streams through independent one-dimensional convolutional feature extractors, followed by (i) a sigmoid gating module that selectively suppresses uninformative background activations while amplifying fall-discriminative features, (ii) a global average pooling layer that compresses each stream into a compact fixed-length descriptor, and (iii) a shared classification head that fuses both descriptors for binary fall prediction. For offline evaluation, we evaluate the model across five wrist-mounted inertial measurement unit (IMU) datasets, achieving average F1-scores of 93%, 93%, 90%, 91%, and 90% on SmartFallMM, WEDA-Fall, FallAllD, UMAFall, and UP-Fall, outperforming Transformer baselines. For real-time evaluation, we deployed the model on a Google Pixel Watch 3 and tested across 12 participants. The model achieves an average F1-score of 97% and an accuracy of 98% with zero missed falls, showing that sigmoid gating offers a more structurally aligned and computationally efficient alternative to attention for commodity smartwatch-based fall detection.

cs.CV↗

arXiv2006

Evaluation of cell wall preparations for proteomics: a new procedure for purifying cell walls from Arabidopsis hypocotyls

The ultimate goal of proteomic analysis of a cell compartment should be the exhaustive identification of resident proteins; excluding proteins from other cell compartments. Plant cell walls possess specific difficulties. Several reported procedures to isolate cell walls for proteomic analyses led to the isolation of a high proportion (more than 50%) of predicted intracellular proteins. The rationales of several published procedures to isolate cell walls for proteomics were analyzed, with regard to the bioinformatic-predicted subcellular localization of the identified proteins. A new procedure was developed to prepare cell walls from etiolated hypocotyls of Arabidopsis thaliana. After salt extraction, a high proportion of proteins predicted to be secreted was released (73%), belonging to the same functional classes as proteins identified using previously described protocols. The new cell wall preparation described in this paper gives the lowest proportion of proteins predicted to be intracellular when compared to available protocols. The application of its principles should lead to a more realistic view of the cell wall proteome, at least for the weakly bound CWP extractable by salts. In addition, it offers a clean cell wall preparation for subsequent extraction of strongly bound CWP.

q-bio.GN↗