Fahad Saeed

CViT-ESP: Lightweight Pre-trained Vision Transformers for EEG-based Epileptic Seizure Prediction

This paper introduces CVIT-ESP and ESPFormer, pre-trained Vision Transformer models adapted for EEG-based seizure prediction, achieving higher AUC-ROC than existing models.

BindScreen: Protein-Centric Contrastive Learning for Sequence-Based Virtual Screening

BindScreen improves virtual screening by using a protein-centric contrastive learning approach with asymmetric loss and batch design, achieving better performance with fewer computational resources.

ProtEnrich: Residual Multimodal Enrichment of Protein Sequence Embeddings

ProtEnrich enhances protein sequence embeddings with biophysical properties without disrupting their semantic structure, improving performance on structure-related tasks.

TITAN-BBB: Predicting BBB Permeability using Multi-Modal Deep-Learning Models

TITAN-BBB predicts BBB permeability using multi-modal deep learning, achieving high accuracy and error reduction compared to existing methods.

MolDeBERTa: Foundational Model for Physicochemical and Substructure-Informed Molecular Representation Learning

MolDeBERTa is a molecular language model pre-trained on SMILES data with novel objectives to improve property prediction and substructure similarity, outperforming existing models on multiple benchmarks.

Habenula alterations in resting state functional connectivity among autistic individuals

The paper investigates habenula functional connectivity differences in autistic individuals compared to neurotypical controls using rs-fMRI data from ABIDE.

A Machine Learning and Benchmarking Approach for Molecular Formula Assignment of Ultra High-Resolution Mass Spectrometry Data from Complex Mixtures

This paper applies machine learning to improve molecular formula assignment in UHRMS data, achieving higher annotation rates and accuracy compared to traditional methods.

FiCOPS: Hardware/Software Co-Design of FPGA Computational Framework for Mass Spectrometry-Based Peptide Database Search

FiCOPS accelerates mass spectrometry peptide database search using FPGA hardware/software co-design, achieving 3.5x speed-up and reduced power consumption.

Utilizing Pretrained Vision Transformers and LargeLanguage Models for Epileptic Seizure Prediction

This paper uses pre-trained Vision Transformers and Large Language Models to predict epileptic seizures, finding that LLMs achieve higher sensitivity than ViTs and a custom ResNet model.

Identifying Alzheimer's disease-associated genes using PhenoGeneRanker

This study uses network biology methods to prioritize potential Alzheimer's disease-associated genes by integrating multimodal datasets and applying PhenoGeneRanker.

Predicting peptide properties from mass spectrometry data using deep attention-based multitask network and uncertainty quantification

ProteoRift predicts peptide properties from mass spectrometry data using a deep attention-based multitask network, reducing search space and improving speed while maintaining accuracy.

MLSPred-Bench: ML-Ready Benchmark LeveragingSeizure Detection EEG data for Predictive Models

The paper converts seizure detection EEG data into ML-ready datasets for predictive modeling, creating MLSPred-Bench with 12 datasets and 150GB of data.

PVTAD: ALZHEIMERS DISEASE DIAGNOSIS USING PYRAMID VISION TRANSFORMER APPLIED TO WHITE MATTER OF T1-WEIGHTED STRUCTURAL MRI DATA

PVTAD classifies Alzheimer's disease and cognitively normal cases using a pyramid vision transformer on white matter MRI data, achieving 97.7% accuracy.

PPAD: A deep learning architecture to predict progression of Alzheimer's disease

This paper introduces two deep learning models, PPAD and PPAD-AE, to predict Alzheimer's disease progression from mild cognitive impairment using EHR data, achieving better performance than baselines in F2 and sensitivity.

SLM-Transform: A Method for Memory-Efficient Indexing of Spectra for Database Search in LC-MS/MS Proteomics

SLM-Transform improves memory efficiency for peptide indexing in LC-MS/MS proteomics by limiting indexed ions and enabling constant-time queries.

MaSS-Simulator: A highly configurable MS/MS simulator for generating test datasets for big data algorithms

MaSS-Simulator generates MS/MS datasets with known ground truth for testing proteomics algorithms, showing high similarity to experimental spectra.