A5001062659- AI-based Problem Solving and Planning
- Constraint Satisfaction and Optimization
- Genomic variations and chromosomal abnormalities
- Software Engineering Research
- Model-Driven Software Engineering Techniques
- Prenatal Screening and Diagnostics
- Machine Learning and Algorithms
- Machine Learning and Data Classification
- Bayesian Modeling and Causal Inference
- Software Testing and Debugging Techniques
- Non-Invasive Vital Sign Monitoring
- Spectroscopy Techniques in Biomedical and Chemical Research
- Distributed and Parallel Computing Systems
- Scheduling and Timetabling Solutions
- Formal Methods in Verification
- Reproductive System and Pregnancy
- Congenital heart defects research
- Genetics and Neurodevelopmental Disorders
- Intelligent Tutoring Systems and Adaptive Learning
- Genetic and Clinical Aspects of Sex Determination and Chromosomal Abnormalities
- German Literature and Culture Studies
- Genomics and Rare Diseases
- Robotic Path Planning Algorithms
- Metaheuristic Optimization Algorithms Research
- Artificial Intelligence in Games
University of British Columbia
2008-2021
Royal Columbian Hospital
2009-2012
BACKGROUNDRecurrent pregnancy loss (RPL), defined as two or more miscarriages, affects 3–5% of couples trying to establish a family. Despite extensive evaluation, no factor is identified in ∼40% cases. In this study, we investigated the possibility that submicroscopic chromosomal changes, not detectable by conventional cytogenetic analysis, exist miscarriages with normal karyotypes (46,XY 46,XX) from idiopathic RPL.
Higher resolution whole-genome arrays facilitate the identification of smaller copy number variations (CNVs) and their integral genes contributing to autism and/or intellectual disability (ASD/ID). Our study describes use one highest arrays, Affymetrix(®) Cytogenetics 2.7M array, coupled with quantitative multiplex polymerase chain reaction (PCR) short fluorescent fragments (QMPSF) for detection validation small CNVs. We studied 82 subjects ASD ID in total (30 52 application cohort) detected...
State-of-the-art planners often exhibit substantial runtime variation, making it useful to be able efficiently predict how long a given planner will take run on instance. In other areas of AI, such needs are met by building so-called empirical performance models (EPMs), statistical derived from sets problem instances and observations. Historically, have been less accurate for predicting the running times planners. A key hurdle has relative weakness in instance features characterizing...
Developmental abnormalities of human embryos can be visualized in utero using embryoscopy. Our previous embryoscopic and genetic evaluations detected developmental the majority both euploid (74%) aneuploid or polyploid (90%) miscarriages. Since we found pattern morphological changes to similar non-euploid embryos, proposed that lethal submicroscopic changes, not by standard chromosome testing, may responsible for miscarriage embryos. Whole genome oligo bacterial artificial array comparative...
To achieve peak performance, it is often necessary to adjust the parameters of a given algorithm class problem instances be solved; this known case for popular solvers broad range AI problems, including planning, propositional satisfiability (SAT) and answer set programming (ASP). avoid tedious highly sub-optimal manual tuning such by means ad-hoc methods, general-purpose configuration procedures can used automatically find performance-optimizing parameter settings. While impressive...
When designing state-of-the-art, domain-independent planning systems, many decisions have to be made with respect the domain analysis or compilation performed during preprocessing, heuristic functions used search, and other features of search algorithm. These design can a large impact on performance resulting planner. By providing alternatives for these choices exposing them as parameters, systems in principle configured work well different domains. However, planners are typically default...
Combining the complementary strengths of several algorithms through portfolio approaches has been demonstrated to be effective in solving a wide range AI problems. Notably, techniques have prominently applied suboptimal (satisficing) planning. Here, we consider construction sequential planner portfolios for domainindependent optimal Specifically, introduce four (three which are dynamic) per-instance schedule generation using problem instance features, and investigate usefulness static...
Solver competitions have been used in many areas of AI to assess the current state art and guide future research development. planning is no exception, International Planning Competition (IPC) has frequently run for nearly two decades. Due organisational comput ational burden involved running these competitions, solvers are generally compared using a single homogeneous hardware software environment all competitors. To what extent does specific choice an effect on solver performance, that...
Combining the complementary strengths of several algorithms through portfolio approaches has been demonstrated to be effective in solving a wide range AI problems. Notably, techniques have prominently applied suboptimal (satisficing) planning. Here, we consider construction sequential planner portfolios for (domain-independent) optimal Specifically, introduce four (three which are dynamic) per-instance schedule generation using problem instance features, and investigate usefulness static...
Putatively benign copy number variants (bCNVs) can be broadly defined as DNA gains or losses that do not lead to a recognizable clinical phenotype. Detection of bCNVs in genomes clinically healthy individuals is increasing with the widespread use whole genome arrays different resolutions and sequence comparison methods. However, role human disease susceptibility phenotype diversity mostly unknown. In order explore potential and/or pathogenesis neurodevelopmental disorders we examined...
Solver competitions play a prominent role in assessing and advancing the state of art for solving many problems AI beyond. Notably, areas AI, have had substantial impact guiding research applications years, solver to be ranked highly competition carries considerable weight. But which extent can we expect results generalise sets problem instances different from those used particular competition? This is question investigate here, using statistical resampling techniques. We show that rankings...
Abstract We describe two males with intellectual disability (ID) and facial dysmorphism, both of whom have non‐mosaic Y chromosome rearrangements resulting in deletions large portions the chromosome. Patient A, ID, mild speech delay, Duane anomaly eye, hypermetropia conductive hearing loss, had structurally rearranged chromosomes p q arm addition to a Yp duplication. B, also language developmental delay short stature, an interstitial deletion Yq11.21–11.23. Array‐CGH excluded presence...
Identifying injuries, deformities, and diseases by non-invasive instrumental means has been a major innovation in medicine. Diagnostic imaging medical devices have revolutionized diagnosis surgery, providing more efficient way to identify injuries diseased or damaged tissues. In this paper, identification of different animal tissues using miniature near-infrared (NIR) spectrometer will be demonstrated. Each tissue type contains amounts moisture proteins, spectrometer, fiber-optic probe...
Modern software systems in many application areas offer to the user a multitude of parameters, switches and other customisation hooks. Humans tend have difficulties determining best configurations for particular applications. optimising compilers are an example such systems; their parameters need be tuned optimal performance, but often left at default values convenience. In this work, we automatically determine compiler parameter settings that result optimised performance Specifically, apply...
Identifying injuries, deformities, and diseases by non-invasive instrumental means has been a major innovation in medicine. Diagnostic imaging medical devices have revolutionized diagnosis surgery, providing more efficient way to identify injuries diseased or damaged tissues. In this paper, identification of different animal tissues using miniature near-infrared (NIR) spectrometer will be demonstrated. Each tissue type contains amounts moisture proteins, spectrometer, fiber-optic probe...