A5101882341- Gut microbiota and health
- Gastric Cancer Management and Outcomes
- Artificial Intelligence in Healthcare
- Esophageal Cancer Research and Treatment
- Liver Disease Diagnosis and Treatment
- Organ Transplantation Techniques and Outcomes
- AI in cancer detection
- Dental Education, Practice, Research
- Colorectal Cancer Screening and Detection
- Global Health Workforce Issues
- Diet and metabolism studies
- Gene expression and cancer classification
- Urological Disorders and Treatments
- Brain Tumor Detection and Classification
- Liver Disease and Transplantation
- Machine Learning in Healthcare
- Radiomics and Machine Learning in Medical Imaging
- Genital Health and Disease
- Oral Health Pathology and Treatment
- Sexual function and dysfunction studies
- Microbial Community Ecology and Physiology
- Multiple and Secondary Primary Cancers
- Ureteral procedures and complications
- Pentecostalism and Christianity Studies
- Forecasting Techniques and Applications
Princess Margaret Cancer Centre
2021-2025
York University
2024-2025
University Health Network
2021-2025
University of Toronto
2020-2025
Public Health Ontario
2020-2025
Toronto General Hospital
2022-2024
Epsom and St Helier University Hospitals NHS Trust
2018-2020
University College London Hospitals NHS Foundation Trust
2020
University College London
2020
Guru Nanak Dev University
2020
Research supports the potential use of microbiome as a predictor some diseases. Motivated by findings that data is complex in nature, and there an inherent correlation due to hierarchical taxonomy microbial Operational Taxonomic Units (OTUs), we propose novel machine learning method incorporating stratified approach group OTUs into phylum clusters. Convolutional Neural Networks (CNNs) were used train within each clusters individually. Further, through ensemble approach, features obtained...
Complex biological features such as the human microbiome and gene expressions play a crucial role in health by mediating various biomedical processes that influence disease progression, immune responses metabolic processes. Understanding these mediation roles is essential for gaining insights into pathogenesis improving treatment outcomes. However, analyzing high-dimensional presents challenges due to their inherent structural correlations, hierarchical taxonomic structures microbial...
Abstract Motivation Research shows that human microbiome is highly dynamic on longitudinal timescales, changing dynamically with diet, or due to medical interventions. In this article, we propose a novel deep learning framework ‘phyLoSTM’, using combination of Convolutional Neural Networks and Long Short Term Memory (LSTM) for feature extraction analysis temporal dependency in sequencing data along host’s environmental factors disease prediction. Additional novelty terms handling variable...
Abstract Background and Aims Bacterial species microbial pathways along with metabolites clinical parameters may interact to contribute non‐alcoholic fatty liver disease (NAFLD) severity. We used integrated machine learning models a cross‐validation approach assess this interaction in bariatric patients. Methods 113 patients undergoing surgery had biochemical parameters, blood stool metabolite measurements as well faecal shotgun metagenome sequencing profile the intestinal microbiome. Liver...
Abstract Motivation Research is improving our understanding of how the microbiome interacts with human body and its impact on health. Existing machine learning methods have shown great potential in discriminating healthy from diseased states. However, Machine Learning based prediction using data has challenges such as, small sample size, imbalance between cases controls high cost collecting large number samples. To address these challenges, we propose a deep framework phylaGAN to augment...
371 Background: The therapeutic approach in locally advanced GEC is evolving. NA-CRT followed by surgery has been recommended based on the CROSS trial. SANO trial reports active surveillance may be an alternative clinically complete response after D-CRT both squamous cell carcinoma (SCC) and adenocarcinoma (AC). SCC AC are different diseases with differences tumor location, etiology, biology. Despite evolving landscape, there still a role for approaches. In our large observational cohort...
Human epidermal growth factor receptor 2 (HER2) overexpression is present in approximately 20-25 of patients with advanced gastroesophageal adenocarcinoma (GEA). Upon progression on 1st line therapy, ramucirumab and paclitaxel (rampac) given ≥2 setting regardless HER2 status. We aim to assess whether associated better survival positive(+) pts compared those HER2(-) disease. reviewed all consecutive adult metastatic/unresectable GEA who were treated rampac for ≥2nd therapy at Princess...
We describe the use of contrast-enhanced ultrasound as an additional imaging technique during examination a traumatised testis, allowing for confident testicular preserving surgery to be performed.
To assess the effects of faecal microbial transplant (FMT) from lean people to subjects with obesity via colonoscopy.In a double-blind, randomized controlled trial, body mass index ≥ 35 kg/m2 and insulin resistance were randomized, in 1:1 ratio blocks four, either allogenic (from healthy donor; n = 15) or autologous FMT (their own stool; 13) delivered caecum followed for 3 months. The main outcome was homeostatic model assessment (HOMA-IR) secondary outcomes glycated haemoglobin levels,...
This paper proposes a hardware model that provides new fire detection and control mechanism with the interface of artificial neural network fuzzy logic.This work is based on integration module implementation inference system (ANFIS).The consists temperature sensor, smoke flame detector microcontroller unit.The sensors sense environment send data to for further processing.Here will as also GSM sending warning message if severe exists, GPS in order indicate location.This technique expresses...
Over the last few years there has been tremendous growth in field of healthcare monitoring systems hospitals and outside it.Developing wireless health care devices employing various technologies become a keen area interest India as well other Nations.This proposed work aims to integrate artificial neural intelligence domain monitoring.Wireless body sensor have ability reach an advance level human utilizing transmission data analytics techniques.Implementation Artificial Neural Fuzzy...
Abstract Background Patient-reported outcomes measures (PROM) are self-reflections of an individual’s physical functioning and emotional well-being. The Edmonton Symptom Assessment Scale (ESAS) is a simple validated PRO tool 10 common symptoms patient-reported functional status (PRFS) measure. prognostic value this unknown in patients with gastroesophageal cancer (GEC). In study, we examined the association between ESAS score overall survival (OS) GEC, prognostication difference Eastern...
ABSTRACT Background and Aims Liver transplant recipients (LTRs) are at risk of developing graft injury, leading to cirrhosis reduced survival. biopsy remains the gold standard method for diagnosis pathology but is invasive risky. Our study aimed develop a novel hybrid multi-class neural network (NN) model ‘GraftIQ’ integrating clinician expertise non-invasive pathology. Methods Graft injury was based on liver biopsies from LTRs (1992-2020). Demographic, clinical, laboratory data 30 days...
Background: High mortality associated with cancer presents challenges in both diagnosis and treatment. Deep learning-based clustering analysis provides a powerful tool for identifying molecular subtypes within enabling personalized treatment strategies. Objectives: This project has two objectives: (1) to cluster gene expression data patients into unique uncover novel biological insights improve patient outcomes; (2) identify signatures driving each cluster. Methods: GDC Pan-Cancer data,...
Importance: Transplantation is one of the few areas in medicine where definitive treatment rationed. Subjective decision-making pose challenges towards transplant selection process. It has been proposed that large language models (LLMs) as artificial intelligent (AI) agents could provide objectivity to solve complex problems. Objective: To examine performance a multidisciplinary committee AI (AI-SC) proof-of-concept liver (LT) Design: The AI-SC consisted four LLMs: hepatologist, surgeon,...
The microbiome is increasingly regarded as a key component of human health, and analysis data can aid in the development precision medicine. Due to high cost shotgun metagenomic sequencing (SM-seq), analyses be done cost-effectively two phases: Phase 1-sequencing 16S ribosomal RNA, 2-SM-seq an informative subsample. Existing research suggests strategies select subsample based on biological diversity dissimilarity metrics calculated using operational taxonomic units (OTUs). However, field has...
Objective: To determine those patient groupings, based on volume and risk, whose optimal urethral reconstructive management might be provided by a reorganisation of UK surgeons. Methods: Between 2010 2017, ~689 men/year were enrolled onto an online audit platform collecting data about reconstruction in the UK; this accrual was compared against hospital episode statistics (HES). The available workforce, where based, collected. Individual institutional incumbent volumes, pathology, surgical...