Reuel A. Stallones Building in the Texas Medical Center in Houston
At six campuses across Texas, The University of Texas Health Science Center at Houston (UTHealth) School of Public Health works to improve the state of public health in Texas every day. Each of our campuses is strategically placed to meet the public health education and research needs of the diverse populations across Texas. UTHealth School of Public Health is the only school of public health in the nation with regional campuses.
The main campus, located in the heart of Houston’s Texas Medical Center, offers students unmatched opportunities for research and employment. The School of Public Health’s five regional campuses are in Austin, Brownville, Dallas, El Paso and San Antonio. Each campus has its own faculty and research specialties. Students can attend class at any of the six campuses via Interactive Television (ITV).
UTHealth School of Public Health is one of six schools of The University of Texas Health Science Center at Houston (UTHealth), the most comprehensive academic health system in The University of Texas System and the U.S. Gulf Coast region. In addition to the School of Public Health, UTHealth is home to schools of biomedical informatics, biomedical sciences, dentistry, medicine and nursing. It also includes a psychiatric hospital, multiple institutes and centers, a growing network of clinics and outreach programs in education and care throughout the region.
The School of Public Health is accredited by the Council on Education for Public Health (CEPH) and the university is accredited by the Southern Association of Colleges and Schools (SACS).
Data Science Certificate This certificate is intended for professionals working in health care or industries related to public health research and biostatistics, and consists of coursework in data science, data analytics and predictions, analytic methods, and data management.
Advanced Data Science Certificate The Advanced Data Science certificate is designed for both students and working professionals who intend to elevate their knowledge and skill-set regarding data science processes and their application. This certificate is an extension of the established Data Science Certificate and consists of learning modules in data science crafted to meet the needs of students, employers, and community partners. Topics include data mining, data science computing, and programming in Python and R
Tree-based methods (Trees, Bagging, Random Forests, Boosting)
Support Vector Machines
Neural Networks
Unsupervised Learning (Clustering Methods)
PH1977: Data Science Computing
This course is an introduction to mainstream programming and high performance computing techniques and tools in data science.
Syllabus-PH1977.pdf
Topics
The fundamental Computer System
Profiling to find bottlenecks
Matrix and Vector computation
Concurrency & Multiprocessing modules
High Performance Computing (using Clusters & GPUs)
Introduction to big data files, text data, and web scraping
Data manipulation and management via Bash, Pandas, Relational Data Management Systems
Introduction to big data systems: Hadoop, Spark
Introduction to convex optimization
Introduction to deep learning via Pytorch
Introduction to (interactive) visualization
PH1978: Machine Learning in Practice
This course is an advanced data analysis and prediction techniques and tools with applications.
Syllabus-PH1978.pdf
Topics
Fundamental concepts of machine learning and its applications.
Overview of Supervised and Unsupervised Learning
Data representation and features engineering
Model Evaluation and Improvement
Algorithm Chains and Pipelines
Introduction to Deep learning models
Working with Text data
Working with Image data
Introduction to Recommendation Systems
Machine learning applications in health sciences
PH1998 Special Topic: Advanced Data Science Analytic Methods
Learn the state-of-the-art concepts and algorithms in deep learning, generative adversarial networks and their application to imaging, EHR and big omics data analysis.
Topics
Deep neural networks
Deep residual neural networks
Dynamics of output of neural networks
Analytic solutions to deep wide neural networks
Gaussian processes
High dimensional data reduction
Variational autoencoders
Graphic variational autoencoders
Variational inference
Generative adversarial networks (GANs)
Wasserstein and conditional GANs
Deep learning counterfactual representations
PH1998 Special Topic: Big Data in Practice
New concepts of Big Data and Big Data analytic methods. Big Data research projects from different application areas.
Topics
Concepts of Data Science and Big Data
Application Examples Big Data Practice
Data visualization and result summarization skills
Collaboration Skills in Big Data Practice
Communication skills, Teamwork skills
Machine Learning and Predictive Modeling Methods: A Big Data Perspective (Big Volume of data, Big Velocity of data, Big Variety of Data, Big Value of Data)
Leadership skills in Big Data Practice
Practice by participating in one of Big Data projects—Learning by doing
This course is an introduction to mainstream programming and high performance computing techniques and tools in data science. Topics for this course include, The fundamental Computer System, Profiling to find bottlenecks, Matrix and Vector computation, Concurrency & Multiprocessing modules, High Performance Computing (using Clusters & GPUs), Introduction to big data files, text data, and web scraping, Data manipulation and management via Bash, Pandas, Relational Data Management Systems, Introduction to big data systems: (Hadoop, Spark), Introduction to convex optimization, Introduction to deep learning via Pytorch and Introduction to (interactive) visualization.
Advanced Data Science Analytic Methods
In this course, we will introduce the current developments of Big Data analytic methods. In particular, we will train students to learn the state-of-the-art concepts and algorithms in deep learning, generative adversarial networks and their application to imaging, HER and big omics data analysis. The emphasis will be on creative thinking, problem-solving skills, and hand-on data exploration to generate and address important scientific and business questions from variety of complex Big Data theoretic developments and applications. flyer.pdf
Communication, Collaboration and Leadership for Biostatisticians and Data Scientists
Designed for aspiring biostatisticians and data scientists alike, our PH 1998: Communication, Collaboration and Leadership for Biostatisticians and Data Scientists course is designed to help equip students with valuable leadership and communication skills that are essential to every domain of the workforce and complement our education program's curriculum.flyer.pdf
Spatial-Temporal Analysis for Population Health Data
This course is designed for students who wish to analyze spatialtemporal
data for population health. Topics include research
ethics, study design, databases for spatial-temporal population
health data, data retrieving and processing, geocoding using
google map API and census geocoder API with R programming
environment, exploratory data analysis and data visualization for
spatial-temporal data. The course will also introduce a variety of
statistical modeling methods for point-level and area-level
population data, and focus on their application and
interpretation. flyer.pdf
The Biostatistics Department of the UTHealth School of Public Health (SPH) offers graduate studies leading to the Master of Science (MS) and Doctor of Philosophy (PhD) degrees.
Biostatistics is a discipline encompassing the study and development of statistical, mathematical, and computer methods applied to the biological and health sciences. Biostatisticians play a key role in the design, conduct and analysis of research studies of health and disease. There is ample opportunity for experience in consulting and collaborative research. Alumni of the Biostatistics program are prominent in academia, industry and government.
Minors for all degree programs can be selected from the Health Promotion and Behavioral Sciences, Environmental and Occupational Health Sciences, Epidemiology and Disease Control and/or Management, Policy and Community Health Departments. In addition to courses at UTHealth SPH, a wide variety of courses are available through cross registration with other schools and institutions in the Texas Medical Center as well as Rice University and the University of Houston.
The MS program is 36 hours, including 6 hours max of thesis (and practicum), which generally takes two years to complete. The MS program includes a sequence of courses in basic statistical theory and methods, research design and data analysis. All MS students take a minor in a field of application other than biostatistics.
Admission Requirements
Undergraduate degree in statistics, mathematics, computer science, or one of the physical, biological, or social sciences
At least a B average (on a 4.0 system) in prior academic work
The PhD program requires at least two additional years beyond the master level program. All UTHealth SPH PhD students must complete a minimum of 48 credits, including 9 hours of dissertation and practicum . The PhD program emphasizes advanced statistical theory and application, statistical consulting, and independent research. All PhD students are required to choose one minor and one area of breadth in fields of application other than biostatistics.
Admission Requirements
Bachelor’s degree in mathematics or statistics or MS degree in the theory and applications of biostatistics, mathematics, statistics or equivalent is required
The division also offers a minor course of study for MS, DrPH and PhD students majoring in other public health disciplines. Courses for the minor of a MS student, include at least PH 1700 (Intermediate Biostatistics) and two biostatistics courses above PH1700. Courses for the minor of a PhD or DrPH student, include at least PH 1700 (Intermediate Biostatistics) and three biostatistics courses above PH1700. PH 1820 (Applied Statistical Analysis I) is strongly recommended for the minor for all degree program.
Data Science Certificate This certificate is intended for professionals working in health care or industries related to public health research and biostatistics, and consists of coursework in data science, data analytics and predictions, analytic methods, and data management.
Advanced Data Science Certificate The Advanced Data Science certificate is designed for both students and working professionals who intend to elevate their knowledge and skill-set regarding data science processes and their application. This certificate is an extension of the established Data Science Certificate and consists of learning modules in data science crafted to meet the needs of students, employers, and community partners. Topics include data mining, data science computing, and programming in Python and R
MS-B1. Use appropriate statistical methods and models to analyze data from the public health, biomedical, or bioinformatics arena.
MS-B2. Demonstrate the correct use of probability distributions and theory of statistical inference within biostatistics and public health.
MS-B3. Outline a statistical analysis strategy to appropriately answer a research question.
MS-B4. Use multiple statistical software packages to analyze data to answer public health research questions.
Doctor of Philosophy (PhD)
Doctor of Philosophy in Biostatistics Competencies
PhD-B1. Prove or derive a statistical theory and apply the result to public health, biomedical, or bioinformatics data.
PhD-B2. Design a simulation study to show or evaluate a method’s effectiveness.
PhD-B3. Use a unified methodological and theoretical framework to analyze various types of data (such as binary, count or continuous data).
PhD-B4. Use advanced computing techniques (for example, EM algorithm and Monte Carlo integration) to analyze complex data.
Guidelines for Doctoral Qualifying Exam
Guidelines for the new doctoral exam, as approved by the Associate Dean for Academic Affairs.
All doctoral students entering in August 2010 or after will follow this new policy and take the new preliminary exam, oral defense and dissertation defense. The Department of Biostatistics preliminary exam will include both take-home and in-class portions. The take-home portion will typically focus on material covered in PH 1820, PH1821, PH1830 and PH1831 and is a two-day exam. The in-class exam will focus on material covered in PH1910, PH1911, PH 1915 and PH1951, is a four-hour exam and students are allowed to bring text books and lecture notes pertinent to the recommended courses listed above. All doctoral students entering before August 2010 can choose either the new system or the old system. If you choose the new system, all rules on the new system should be followed. If you choose the old system, you will take the qualifying exam after completing 30 credit hours and demonstrating that you fulfill the requirements of the minor and the breadth. However, if your qualifying exam takes place after January 2011, your biostatistics portion will be the same format as the preliminary exam according to the new system. January 2011 will be the last time the department will give the old format of qualifying exam for biostatistics students.
Those under the old exam system are expected to take biostatistics courses suggested by your advisory committee to meet the requirements of the degree program, regardless of which biostatistics exam you will take. Those under the new exam system are still expected to take additional biostatistics courses (beyond those recommended above) as well as the minor and breadth courses after the preliminary exam as you develop your proposal under the supervision of your dissertation committee.
The biostatistics preliminary exam will be given once a year in August.
It is the School’s policy that a student who failed twice on preliminary exam (new system) or qualifying exam (old system) can not stay in the doctoral program.
Addendum to PhD Students-Teaching Requirements
Doctoral students are required to obtain some teaching experience on biostatistics courses for majors for at least one semester. A typical example is to serve as a teaching assistant for a high level course in biostatistics after they complete the preliminary exam.
Requirements for Thesis/Dissertation
The dissertation should be in a paper format and is supposed to include two submitted papers. A dissertation proposal defense is required before the student advances to doctoral candidacy. For submitted papers, only the dissertation chair needs to verify. For further information about Dissertation and Thesis Proposal.
Preliminary (Qualifying) Exam Awards
All the Biostatistics and Data Science PhD students who pass the preliminary (qualifying) exam on their first attempt are eligible for the “Qualifying Exam Awards”:
Best Preliminary-Exam Award: Highest final score
Best Preliminary-Exam Award for Statistical Theories: Highest theory score
Best Preliminary-Exam Award for Statistical Applications: Highest application score
One student may get one or more of the three awards. The awardees will receive an award certificate and will be considered a high priority for TA or GRA positions if the positions are available and the awardees are eligible.
Career
Career opportunities abound in the field of biostatistics throughout academia, industry and government. Examples include the pharmaceutical industry, the chemical industry, medical research centers, schools of public health, medical schools and government agencies such as the National Institutes of Health, Centers for Disease Control and Prevention, National Center for Health Statistics, state and local health departments, and the World Health Organization
Faculty members in the Department of Biostatistics & Data Science have led and contributed to the development of cutting-edge statistical and data science methods for many areas including clinical trial design and analysis, longitudinal and correlated data analysis, machine/statistical learning approaches, survival data analysis, imaging data analysis, Bayesian statistics, statistical genomics and genetics, bioinformatics, graphic and network modeling, stochastic processes, missing data, time series and streaming data, spatial-temporal data, dynamic modeling and prediction, and Big Data approaches. These methods are applied to a wide range of biomedical and health science problems including:
Clinical trials for cardiovascular cell therapy, hypertension, stem cells, early treatment of retinopathy of prematurity, heart attack prevention, early treatment of acute spinal cord injury
Alzheimer and Parkinson’s diseases
Brain injuries and neurosciences
Behavioral sciences and mental health
Infectious diseases: HIV prevention and AIDS treatment
Occupational and environmental exposures in the etiology of adult leukemia
Inter-uterine growth through ultrasound measurement
Cancer and cervical cancer detection using optical spectroscopy,
Analysis of infant mortality in developing countries
Queuing models for emergency medical services
Stochastic modeling of movement through the health care system
Health effects of air pollution
Analysis of health services utilization and health care technology assessment
US-Mexico border health issues
Molecular evolution and phylogenetics
Electronic health records (EHR) and medical insurance claim data
Assistant/Associate Professor of Instruction (NTT)
Department of Biostatistics & Data Science at School of Public Health is an academic home for biostatisticians and data scientists at UTHealth to conduct high-quality education, research and service to improve the health of the people of Texas, the nation and the world. Our Department offers the following education programs:
Data Science Certificates
MS/PhD in Public Health with a major in Biostatistics
Our faculty members have a diverse array of statistical and data science methodology expertise with application interests in variety of biomedical and health science fields. For more details, please see our faculty profile webpage
Our Department hosts three research and collaboration/service centers:
Coordinating Center for Clinical Trials (CCCT) which has a more-than-40-year history of successfully coordinating large multicenter clinical trials and providing comprehensive services for clinical trial operation, data management and data analysis.
Center for Big Data in Health Sciences (CBD-HS) hosts several Big Data research projects, including development of predictive models and heterogeneous data integration methods for Electronic Healthcare Record (EHR) data, insurance claim data and public databases or data repositories.
Center for Biostatistics Collaboration and Data Services will provide opportunities for our faculty and students to collaborate and interact with biomedical and health science investigators to address scientific questions using their statistical and data science skills.
