Andrew Lee
A5100450498- BRCA gene mutations in cancer
- Natural Language Processing Techniques
- Advanced Malware Detection Techniques
- Topic Modeling
- Target Tracking and Data Fusion in Sensor Networks
- Speech and dialogue systems
- Network Security and Intrusion Detection
- Semantic Web and Ontologies
- Genetic Associations and Epidemiology
- Ovarian cancer diagnosis and treatment
- Structural Health Monitoring Techniques
- Inertial Sensor and Navigation
- Fault Detection and Control Systems
- Cybercrime and Law Enforcement Studies
- Genetic factors in colorectal cancer
- Robotic Path Planning Algorithms
- Anomaly Detection Techniques and Applications
- Nutrition, Genetics, and Disease
- Robotic Mechanisms and Dynamics
- Genomic variations and chromosomal abnormalities
- Information and Cyber Security
- Machine Learning in Healthcare
- Scientific Research and Philosophical Inquiry
- Robotics and Sensor-Based Localization
- Adaptive Control of Nonlinear Systems
St. Thomas Aquinas College
2023-2025
University of California, Berkeley
2011-2024
Google (United States)
2024
DeepMind (United Kingdom)
2024
Rutgers, The State University of New Jersey
2024
Rutgers New Jersey Medical School
2024
McMaster University
2023
University of Guelph
2017-2023
University of Michigan
2019-2023
York College of Pennsylvania
2023
PURPOSE To estimate age-specific relative and absolute cancer risks of breast to ovarian, pancreatic, male breast, prostate, colorectal cancers associated with germline PALB2 pathogenic variants (PVs) because these have not been extensively characterized. METHODS We analyzed data from 524 families PVs 21 countries. Complex segregation analysis was used (RRs; country-specific population incidences) cancers. The models allowed for residual familial aggregation ovarian were adjusted the...
Genome-wide association studies (GWAS) have identified 94 common single-nucleotide polymorphisms (SNPs) associated with breast cancer (BC) risk and 18 ovarian (OC) risk. Several of these are also BC or OC for women who carry a pathogenic mutation in the high-risk genes BRCA1 BRCA2. The combined effects variants on BRCA2 carriers not yet been assessed while their clinical management could benefit from improved personalized estimates. We constructed polygenic scores (PRS) using susceptibility...
CHEK2*1100delC is a well-established breast cancer risk variant that most prevalent in European populations; however, there are limited data on of by age and tumor subtype, which limits its usefulness prediction. We aimed to generate subtype- age-specific estimates using from the Breast Cancer Association Consortium, including 44,777 patients with 42,997 controls 33 studies genotyped for CHEK2*1100delC.
The purpose of this study was to estimate precise age-specific tubo-ovarian carcinoma (TOC) and breast cancer (BC) risks for carriers pathogenic variants in RAD51C RAD51D.
Abstract Introduction Previous studies have demonstrated that common breast cancer susceptibility alleles are differentially associated with risk for BRCA1 and/or BRCA2 mutation carriers. It is currently unknown how these different subtypes in and carriers defined by estrogen (ER) or progesterone receptor (PR) status of the tumour. Methods We used genotype data on up to 11,421 7,080 carriers, whom 4,310 had been affected information either ER PR tumour, assess associations 12 loci tumour...
Abstract Introduction Several common alleles have been shown to be associated with breast and/or ovarian cancer risk for BRCA1 and BRCA2 mutation carriers. Recent genome-wide association studies of identified eight additional susceptibility loci: rs1011970 (9p21, CDKN2A/B) , rs10995190 ( ZNF365) rs704010 ZMIZ1 ), rs2380205 (10p15), rs614367 (11q13), rs1292011 (12q24), rs10771399 (12p11 near PTHLH) rs865686 (9q31.2). Methods To evaluate whether these single nucleotide polymorphisms (SNPs) are...
Abstract This study investigated the speech recognition abilities of popular voice assistants when being verbally asked about commonly dispensed medications by a variety participants. Voice recordings 46 participants (12 which had foreign accent in English) were played back to Amazon’s Alexa, Google Assistant, and Apple’s Siri for brand- generic names top 50 most United States. A repeated measures ANOVA indicated that Assistant achieved highest comprehension accuracy both brand medication (...
Background Genome-wide association studies have identified >30 common SNPs associated with epithelial ovarian cancer (EOC). We evaluated the combined effects of EOC susceptibility on predicting risk in an independent prospective cohort study. Methods genotyped single nucleotide polymorphisms (SNPs) a nested case–control study (750 cases and 1428 controls) from UK Collaborative Trial Ovarian Cancer Screening trial. Polygenic scores (PRSs) were constructed their associations using logistic...
Background The currently known breast cancer-associated single nucleotide polymorphisms (SNPs) are presently not used to guide clinical management. We explored whether a genetic test that incorporates SNP-based polygenic risk score (PRS) is clinically meaningful in non- BRCA1/2 high-risk cancer families. Methods 101 families were included; 323 cases and 262 unaffected female relatives genotyped. 161-SNP PRS was calculated standardised 327 population controls (sPRS). Association analysis...
In this paper, an adaptive formulation of the sliding innovation filter (SIF) is presented. The SIF a recently proposed estimation strategy that has demonstrated robustness to modeling errors and uncertainties. It utilizes switching gain function (measurement error) boundary layer term. time-varying derived based on minimizing state error covariance. resulting solution creates SIF. applied linear aerospace system, compared with well-known Kalman (KF) standard results demonstrate new in...
Germline mutations in BRCA1 and BRCA2 are associated with increased risks of breast ovarian cancer. A genome-wide association study (GWAS) identified six alleles risk cancer for women the general population. We evaluated four these loci as potential modifiers mutation carriers. Four single-nucleotide polymorphisms (SNPs), rs10088218 (at 8q24), rs2665390 3q25), rs717852 2q31), rs9303542 17q21), were genotyped 12,599 7,132 carriers, including 2,678 cases. Associations within a retrospective...
Stefan Larson, Anish Mahendran, Andrew Lee, Jonathan K. Kummerfeld, Parker Hill, Michael A. Laurenzano, Johann Hauswald, Lingjia Tang, Jason Mars. Proceedings of the 2019 Conference North American Chapter Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers). 2019.
Unselected population-based personalised ovarian cancer (OC) risk assessment combining genetic/epidemiology/hormonal data has not previously been undertaken. We aimed to perform a feasibility study of OC stratification general population women using tool followed by management. Volunteers were recruited through London primary care networks. Inclusion criteria: ≥18 years. Exclusion prior ovarian/tubal/peritoneal cancer, previous genetic testing for genes. Participants accessed an...
Recent progress in large language models (LLMs) has enabled the deployment of many generative NLP applications. At same time, it also led to a misleading public discourse that ``it's all been solved.'' Not surprisingly, this has, turn, made researchers -- especially those at beginning their careers worry about what research area they should focus on. Has solved, or remaining questions can we work on regardless LLMs? To address question, paper compiles directions rich for exploration. We...
The smooth variable structure filter (SVSF) has seen significant development and research activity in recent years. It is based on sliding mode concepts, which utilize a switching gain that brings an inherent amount of stability to the estimation process. In effort improve upon numerical SVSF, square-root formulation derived. SVSF Potter’s algorithm. proposed computationally more efficient reduces risks failure due instability. new strategy applied target tracking scenarios for purposes...
Many statistical models have high accuracy on test benchmarks, but are not explainable, struggle in low-resource scenarios, cannot be reused for multiple tasks, and easily integrate domain expertise. These factors limit their use, particularly settings such as mental health, where it is difficult to annotate datasets model outputs significant impact. We introduce a micromodel architecture address these challenges. Our approach allows researchers build interpretable representations that embed...
This work proposes a new learning-to-search benchmark and uses AI to discover mathematical knowledge related an open conjecture of Erdos (1975) in extremal graph theory. The problem is find graphs with given size (number nodes) that maximize the number edges without having 3- or 4-cycles. We formulate this as sequential decision-making compare AlphaZero, neural network-guided tree search, tabu heuristic local search method. Using either method, by introducing curriculum---jump-starting for...