Student Project Examples
Student Projects
Students had the opportunity to be engaged in a variety of research projects. The projects centered around the following topics: Stony Brook University Covid-19 Positive Cases; A Lareg-Scale CT & PET/CT Dataset Test for Lunch Cancer Diagnosis; Framingham Heart Study Data Extract on Smoking & Hypertension; Using Data to Investigate Elephant Evolution; Evolution in Action: Data Analysis; HEad & Neck PET-CT-Cancer Dataset; Investigation Planarian Behavior & Regeneration.
In each project, model building and data analysis played a critical role and was interwoven in a statistical and biological context. Listed below is a brief description of each project as well as the names of students involved in the research. The students reported their research findings to their parents and university faculty on the last day of the Governor’s School.
Project #1
("Stony Brook University Covid-19 Positive Cases") - There were five students who participated in this project.
This collection of cases was acquired at Stony Brook University from patients who tested positive for COVID-19. The collection includes images from different modalities and organ sites (chest radio[1]graphs, chest CTs, brain MRIs, etc.).
https://wiki.cancerimagingarchive.net/pages/ viewpage.action?pageId=89096912
Project #2
("A Large-Scale CT & PET/CT Dataset for Lung Cancer Diagnosis") - There were five students who participated in this project.
The data were obtained from the OECD data base. This dataset consists of CT and PET-CT DICOM images of lung cancer subjects with XML Annotation files that indicate tumor location with bounding boxes. The images were retrospectively acquired from patients with suspicion of lung cancer, and who under-went standard-of-care lung biopsy and PET/CT. Subjects were grouped according to a tissue histopathological diagnosis.
https://wiki.cancerimagingarchive.net/pages/ viewpage.action?pageId=70224216
Project #3
("Framingham Heart Study Data Extract on Smoking & Hypertension") - There were five students who participated in this project.
Data collected over a 24-year study suitable for competing risks survival analysis of hypertension and death as a function of smoking.
https://www.rdocumentation.org/packages/ LocalControl/versions/1.1.2.2/topics/framingham/
Project #4
("Lysine & Energy Content in Livestock Feed") - There were five students who participated in this project.
Hogs were fed diets that varied in lysine levels and energy content over 9 months. Response variables such as feed per gain (FPG), average daily gain (ADG), and average daily feed (ADF) were recorded for each animal during the final month. Grouping variables such as animal sex and barn location were also recorded in addition to lysine levels and energy content and can be used in an analysis, but the project should focus on the impact that lysine levels and energy content had on response variables. Data will be provided by Dr. Chapman .
Project #5
("Gene Expression & Transcription Factors") - There were five students who participated in this project.
Here, you will create your own dataset based on a question (or multiple questions) about how positive and/or negative transcription factors and their affinities influence gene expression. These data will be collected using a simulation on PhET from University of Colorado (link below, when follow, use “mRNA” simulation). Come up with experiments based on varying levels of a given variable (i.e., low, medium and high transcription factors at one level of affinity), and run the simulation in each case for 3 minutes. Count the number of times that the gene is transcribed during that time and record it. Do this several times for each case and record your data for analysis. Don’t hesitate to discuss your experiment and data collection with Dr. Chapman and/or Dr. Price!
Project #6
("Head & Neck PET-CT-Cancer Dataset") - There were five students who participated in this project.
This collection contains FDG-PET/CT and radiotherapy planning CT imaging data of 298 patients from four different institutions in Qu´ebec with histologically proven headand-neck cancer (H&N) All patients had pre-treatment FDGPET/CT scans between April 2006 and November 2014, and within a median of 18 days (range: 6-66) before treatment.
https://wiki.cancerimagingarchive.net/display/Public/HeadNeck-PET-CT