Alzheimer’s disease diagnosis from single and multimodal data using machine and deep learning models: Achievements and future directions

Expert Systems with Applications, Journal Year: 2024, Volume and Issue: 255, P. 124780 - 124780

Published: July 14, 2024

Language: Английский

---

Advance in peptide-based drug development: delivery platforms, therapeutics and vaccines

Signal Transduction and Targeted Therapy, Journal Year: 2025, Volume and Issue: 10(1)

Published: March 5, 2025

The successful approval of peptide-based drugs can be attributed to a collaborative effort across multiple disciplines. integration novel drug design and synthesis techniques, display library technology, delivery systems, bioengineering advancements, artificial intelligence have significantly expedited the development groundbreaking drugs, effectively addressing obstacles associated with their character, such as rapid clearance degradation, necessitating subcutaneous injection leading increasing patient discomfort, ultimately advancing translational research efforts. Peptides are presently employed in management diagnosis diverse array medical conditions, diabetes mellitus, weight loss, oncology, rare diseases, additionally garnering interest facilitating targeted platforms advancement vaccines. This paper provides an overview present market clinical trial progress therapeutics, platforms, It examines key areas through literature analysis emphasizes structural modification principles well recent advancements screening, design, technologies. accelerated including peptide-drug complexes, new vaccines, innovative diagnostic reagents, has potential promote era precise customization disease therapeutic schedule.

Language: Английский

---

A machine learning approach to leveraging electronic health records for enhanced omics analysis

Nature Machine Intelligence, Journal Year: 2025, Volume and Issue: 7(2), P. 293 - 306

Published: Jan. 16, 2025

Omics studies produce a large number of measurements, enabling the development, validation and interpretation systems-level biological models. Large cohorts are required to power these complex models; yet, cohort size remains limited due clinical budgetary constraints. We introduce omics multimodal analysis enhanced with transfer learning (COMET), machine framework that incorporates large, observational electronic health record databases improve small datasets from studies. By pretraining on data adaptively blending both early late fusion strategies, COMET overcomes limitations existing methods. Using two independent datasets, we showed improved predictive modelling performance discovery compared traditional incorporating into analyses, enables more precise patient classifications, beyond simplistic binary reduction cases controls. This can be broadly applied reveals powerful insights sizes.

Language: Английский

---

A systematic review of intermediate fusion in multimodal deep learning for biomedical applications

Image and Vision Computing, Journal Year: 2025, Volume and Issue: unknown, P. 105509 - 105509

Published: March 1, 2025

Language: Английский

---

Open challenges and opportunities in federated foundation models towards biomedical healthcare

BioData Mining, Journal Year: 2025, Volume and Issue: 18(1)

Published: Jan. 4, 2025

This survey explores the transformative impact of foundation models (FMs) in artificial intelligence, focusing on their integration with federated learning (FL) biomedical research. Foundation such as ChatGPT, LLaMa, and CLIP, which are trained vast datasets through methods including unsupervised pretraining, self-supervised learning, instructed fine-tuning, reinforcement from human feedback, represent significant advancements machine learning. These models, ability to generate coherent text realistic images, crucial for applications that require processing diverse data forms clinical reports, diagnostic multimodal patient interactions. The incorporation FL these sophisticated presents a promising strategy harness analytical power while safeguarding privacy sensitive medical data. approach not only enhances capabilities FMs diagnostics personalized treatment but also addresses critical concerns about security healthcare. reviews current settings, underscores challenges, identifies future research directions scaling FMs, managing diversity, enhancing communication efficiency within frameworks. objective is encourage further into combined potential FL, laying groundwork healthcare innovations.

Language: Английский

---

An overview of methods and techniques in multimodal data fusion with application to healthcare

International Journal of Data Science and Analytics, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 10, 2025

Language: Английский

---

ADMNet: adaptive deformable convolution large model combining multi-level progressive fusion for Building Change Detection

Geo-spatial Information Science, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 18

Published: Jan. 17, 2025

Building Change Detection (BCD) based on high-resolution Remote Sensing Images (RSI) simplifies urban surface monitoring. Nevertheless, the mainstream detection methods utilizing traditional convolution and attention mechanisms are often prone to errors due loss of edge detail information underutilization global context information. To address these issues, this paper presents a large model, namely ADMNet, which is built adaptive deformable designed handles various types building change First, we propose Siamese neural network (ADC) modules. The ADC module incorporates spatial offset parameters into convolutional kernel sampling mapping weights capture irregularly varying features for local receptive fields. Second, utilize model semantically driven enhance awareness construct long-range feature dependencies from multi-scale information, then integrated with locally structure achieve accurate localization. Furthermore, design Multi-Level Progressive Feature Fusion (MLPFF) that enhances characterization capabilities ensure internal integrity improves performance by integrating priori knowledge large-model transfer learning. evaluate effectiveness generalizability conduct comparative experiments current two datasets, LEVIR-CD WHU-CD, land cover dataset, SYSU-CD. results show ADMNet outperforms all methods. source code available at https://github.com/spaceYu180/ADMNet.

Language: Английский

---

Surveying the deep: A review of computer vision in the benthos

Ecological Informatics, Journal Year: 2025, Volume and Issue: unknown, P. 102989 - 102989

Published: Jan. 1, 2025

Language: Английский

---

Diagnosis of brain diseases in fusion of neuroimaging modalities using deep learning: A review

Information Fusion, Journal Year: 2022, Volume and Issue: 93, P. 85 - 117

Published: Dec. 14, 2022

Language: Английский

---

Multimodal attention-based deep learning for Alzheimer’s disease diagnosis

Journal of the American Medical Informatics Association, Journal Year: 2022, Volume and Issue: 29(12), P. 2014 - 2022

Published: Sept. 23, 2022

Abstract Objective Alzheimer’s disease (AD) is the most common neurodegenerative disorder with one of complex pathogeneses, making effective and clinically actionable decision support difficult. The objective this study was to develop a novel multimodal deep learning framework aid medical professionals in AD diagnosis. Materials Methods We present Multimodal Disease Diagnosis (MADDi) accurately detect presence mild cognitive impairment (MCI) from imaging, genetic, clinical data. MADDi that we use cross-modal attention, which captures interactions between modalities—a method not previously explored domain. perform multi-class classification, challenging task considering strong similarities MCI AD. compare previous state-of-the-art models, evaluate importance examine contribution each modality model’s performance. Results classifies MCI, AD, controls 96.88% accuracy on held-out test set. When examining different attention schemes, found combination self-attention performed best, no layers model worst, 7.9% difference F1-scores. Discussion Our experiments underlined structured data help machine models contextualize interpret remaining modalities. Extensive ablation studies showed any mixture input features without access information suffered marked performance losses. Conclusion This demonstrates merit combining multiple modalities via deliver highly accurate diagnostic support.

Language: Английский

---