Experimental results demonstrate that LSEAD provides a practical, secure, and scalable approach for early AD screening, and shows that LLM-based embeddings generalize well across datasets and improve AD classification accuracy by up to 5 percent over existing methods.
Abstract
Early diagnosis of Alzheimer's disease (AD) is critical for enabling timely interventions that may slow disease progression and improve patient outcomes. There is a growing need for AD detection methods that are non-invasive and cost-effective, especially in real-world clinical settings with diverse patient populations and recording conditions. Speech-based screening addresses these needs by using natural speech collected without specialized equipment. Recent advances in large language models (LLMs) have improved speech analysis by providing rich linguistic representations and strong generalization. In this study, we propose LSEAD, a speech-based AD detection framework using pretrained open-source LLMs. Speech recordings are automatically transcribed, and text embeddings are extracted using locally deployed LLMs. Principal component analysis (PCA) is applied to reduce dimensionality before classification. Because the framework relies only on speech transcripts and locally deployed models, it supports privacy-preserving AD risk assessment without external data exchange. We evaluate LSEAD on the ADReSS20 and ADReSSo2021 benchmark datasets. Experimental results show that LLM-based embeddings generalize well across datasets and improve AD classification accuracy by up to 5 percent over existing methods, especially for early-stage detection. These results demonstrate that LSEAD provides a practical, secure, and scalable approach for early AD screening.
Cognitive impairment (CI) is a growing public health concern. Early and accurate diagnosis is critical for enabling timely intervention and improving patient outcomes. Speech-based CI detection has emerged as a promising non-invasive approach, as speech signals encode both linguistic and acoustic markers associated with cognitive decline. Recent advances in large language models (LLMs) further strengthen the potential of speech-based assessment by enabling more expressive representation learning and improved generalization across diverse speakers, recording devices, and clinical environments. Moreover, multimodal learning by jointly modeling linguistic and acoustic features allows for a more comprehensive characterization of cognitive and behavioral changes related to CI, leading to more reliable detection. In this work, we propose a multimodal CI detection framework based on open-source LLMs that integrates speech audio and corresponding transcripts while preserving patient privacy. Acoustic embeddings are extracted directly from speech signals, while textual embeddings are generated from automatically transcribed speech. These modality-specific embeddings are then concatenated to create a combined feature vector and used for downstream classification, without requiring access to raw or sensitive patient data. The proposed approach is evaluated on the ADReSS20 and ADReSSo21 benchmark datasets. Experimental results show that the proposed multimodal framework achieves an CI classification accuracy of 92.4% and consistently outperforms single-modality baselines. Our work establishes a new state-of-the-art for CI identification, with the proposed method demonstrating superior cross-dataset generalization. This advance highlights the power of an LLM-based multimodal framework that fuses linguistic and acoustic data to enable robust, scalable, and non-invasive screening.
Ying-lei Huang, Xin Wang, Yuhan Su et al.· 0 citations
It is still hard to find Alzheimer's disease (AD) early, especially when neuroimaging is expensive or tools that depend on language are not available. Spontaneous speech provides a non-invasive signal; however, numerous current methodologies depend on transcripts/ASR or computationally intensive deep models. We offer a simple, audio-only baseline for detecting AD using 176 Cookie Theft recordings from the DementiaBank Pitt corpus (88 AD, 88 controls). WebRTC voice activity detection (VAD) is used to separate speech from non-speech. We take out 99 hand-crafted acoustic-temporal features, including pause and fluency statistics, spectral/prosodic descriptors, and MFCC summaries with {\Delta} and {\Delta}{\Delta}. Evaluation is performed using a stringent speaker-independent GroupShuffleSplit,documenting performance across 30 iterations. A lightweight SVM with an RBF kernel gets an average AUC of 0.674 across runs. For example, a single split has an AUC of 0.742 and an accuracy of 0.657. We also present an exploratory compact-feature analysis utilizing a Top-20 subset ranked by Random Forest importance; since selection is not nested within training splits, these results may be overly optimistic and are not employed for primary conclusions (AUC 0.719). The results indicate that transcript-free spectro-temporal and fluency-related cues can facilitate speaker-independent Alzheimer's disease screening from raw audio, establishing a practical foundation for deployment-oriented research.
Rashin Gholijani Farahani, A. Bastanfard· 0 citations
Parkinson's Disease (PD) is a progressive neurodegenerative disorder, which also impacts speech and vocal characteristics at the initial stages of the disorder. This study proposes an explainable hybrid Bi-LSTM and XG-Boost for detection Parkinsons Disease using non-invasive voice recording, and is designed to be interpretable. The Bidirectional Long Short-Term Memory (Bi-LSTM) network is employed to learn deep latent representations from vocal biomarkers and the final classification is done by the XG-Boost network using both the deep latent representations and the handcrafted voice features. It includes Explainable Artificial Intelligence (XAI) methods like SHAP and LIME to improve interpretability and transparency. SHAP is to identify the globally important vocal features that affect the model prediction, while LIME gives the explanations of individual predictions by instances. The purpose of this work is not to develop a complete clinical implementation tool, as deployment in healthcare and real world will require a detailed clinical validation and regulatory clearance. Rather, in this research we will develop an explainable and transparent AI framework that can be the basis for future researchers and developers to design robust clinical support systems and healthcare applications. The proposed approach has the potential to incorporate the deep learning, ensemble learning and explainable AI concepts in an interpretable Parkinson's disease prediction.
T. Bhutia, Passang Tamang, Dewash Manger et al.· 2026 7th International Confe...· 0 citations
Recent advances in Natural Language Processing (NLP) have enabled new approaches for assessing cognitive health through speech analysis. Dementia remains a growing public health concern, yet diagnosis often relies on costly and time-consuming clinical assessments. Speechbased screening offers an alternative, as language deterioration is among the earliest signs of cognitive decline. However, most existing methods fail to effectively capture both how something is said (from the audio) and what is said (from the transcribed text). Yet both components are critical for accurate and reliable diagnosis. Therefore, this work explores independent approaches to evaluate the best-performing methods. Through systematic experimentation, it was found that standard NLP preprocessing pipelines remove clinically relevant disfluency patterns from speech transcripts before modelling, discarding features that may distinguish dementia from healthy speech. To address this, the Pause-Augmented for Disfluency Markers framework was developed, a novel preprocessing approach that explicitly encodes speech disfluencies as special tokens, preserving them for consumption by the classifiers.
Parkinson's disease (PD) is the second most common neurodegenerative disorder. Typical machine learning screening methods require PD labels, but the available data is limited by privacy concerns and the need for expert annotation. We propose a label-free face-plus-voice PD screen built entirely on frozen pretrained encoders--a face-expression Vision Transformer and HuBERT--in which no PD label touches any fit; the reference is training controls only. The voice modality uses a synthetic-dysarthria contrastive activation addition (CAA) direction built from time-stretch and breathy degradation of healthy speech; the face modality uses a k-nearest-neighbor anomaly score to the control embedding cluster. We introduce the alignment principle, a post-hoc analysis showing that a synthetic-degradation CAA detector works when the cosine similarity between the synthetic and real disease directions exceeds zero. Measured on the YouTubePD benchmark, this cosine is +0.37 for voice (CAA works, AUROC 0.765) and -0.48 for face (CAA fails; anomaly succeeds, AUROC 0.751). Equal-weight late fusion reaches AUROC 0.802 (95% CI [0.70,0.89]) with NPV 0.95, supporting a rule-out triage interpretation. An overfitting audit shows the voice detector transfers cleanly, while the face-side--and thus fused--AUROC is potentially optimistic pending external validation.
Initial identification of Alzheimer’s disease (AD) a progressive neurological illness can greatly enhance patient care. Alternative approaches are necessary since traditional diagnostic techniques like neuroimaging are expensive, intrusive, and less accessible. This research work reviews the literature of existing studies covering 2018–2025 and provides a thorough literature review analysis on explainable handwriting approaches based on Natural Language Processing (NLP) for the early prediction of Alzheimer’s disease . The review highlights that stroke velocity, pressure variation, spatial consistency and temporal pauses are very important motor biomarkers of cognitive decline in handwriting. it also states that lexical richness, syntactic complexity, and semantic coherence are linguistic features which when examined through NLP driven textual analysis corroborate these findings. When it comes to distinguishing AD from control groups, machine learning and deep learning models such as Random Forest, XGBoost, Convolutional Neural Network (CNN), and transformer-based architectures have reached accuracies above 90%. Significantly, clinical interpretability has been improved with the use of Explainable Artificial Intelligence (XAI) approaches like Gradient-weighted Class Activation Mapping (Grad-CAM), SHapley Additive exPlanations (SHAP), and cal Interpretable Model-agnostic Explanations (LIME), which enable the display of diagnostic reasoning at the feature level. There are still some unanswered questions about standardized explainability frameworks, cross-population generalizability, and dataset heterogeneity, notwithstanding these advances. In conclusion, this analysis highlights the promising future of explainable NLP-handwriting systems as a means to bridge computational innovation with clinical application; these systems might be used as non-invasive, interpretable, and cost-effective diagnostic methods for Alzheimer’s disease.
Aswathy S, K.Vidhya, A. Ebenezer· 2026 4th International Confe...· 0 citations
