Jackson Heart Study
Jackson Heart Study (JHS) project address health disparities of African Americans (AAs) in Mississippi regarding obesity and obesity-related diseases. In this project we are evaluating abdominal CT exams that were performed during Exam-2 on 2883 participants of the JHS which is a large, well-characterized community-based cohort of AAS, with extensive phenotyping, longitudinal surveillance, and adjudication of events. We supervise a group of medical students, residents and faculty member who execute measurements for a proposal to perform the first comprehensive assessment of body composition analysis in relation to cardiometabolic risk and obesity-related diseases in AAs. During the execution of this project, more than 10 collaborators were trained in the use of different softwares such as Osirix, Slice-O-matic, LSN, QCT Pro, etc. through which various parameters of CT abdominal images are extracted.
Intracranial tumor (GBM) response based on RANO criteria
A computational environment (GUI) that allows to determine in a simple, exact and fast way the response of intracranial tumors (e.g. Glioblastoma Multiforme) through the Response Assessment in Neuro-Oncology (RANO) criteria is being developed.
The GUI will have an image assistant that will allow the user to measure the lengths of the evaluated tumor(s). The set of length results will be automatically exported to a visual environment where the user, besides visualizing the tumor(s) and their most significant measurements, will be able to obtain the result of the analysis based on the RANO criteria.
Radiomics predict response/outcome in head & neck cancer patients treated with chemoradiation
Surveillance imaging for patients who are treated with definitive chemoradiation has inherent pitfalls due to the difficulty to differentiate residual disease from radiation changes. These shortcomings in the imaging might lead to unnecessary interventions including salvage surgery, thus resulting in worse Quality of Life for the patients and increased health care expenses. This study assessed the response and progress of tonsillar cancer patients treated with chemoradiation with the purpose of differentiate between residual tumor and radiation changes through radiomics features extracted from pre-treatment and post-treatment CT images.
Quantification of the Cardiac Motion Field in MRI heart images
The progression of arterial hypertension (untreated or poorly treated) to LV dysfunction is a process that initiates without apparent symptoms or warning signs. In this scenario, the quantification of cardiac left ventricular (LV) motion from medical images provides a non-invasive method for effectively diagnosing the presence or absence of LV dysfunction. The cardiac LV motion (contraction and relaxation) can be qualitatively analyzed and quantified from cardiac magnetic resonance (CMR) image sequences throughout the cardiac cycle as a basis for early reliable diagnosis and prognosis with staging of disease.
We are working on the implementation of a diagnostic system (software) to detect segmental and global cardiac abnormalities of contraction and relaxation and quantify the degree of such defects in diastolic and systolic dysfunction based on:
- To generate a cardiac motion map through computational techniques to determine the systolic and diastolic global and segmental capacity of the myocardium (systolic and diastolic function of the LV) resulting in cardiovascular risk.
- To characterize the morphological changes of normal LV regions respect to the region with decreased systolic relaxation and/or contractile function through extraction of radiomic features based on CMR imaging.
- To quantify the metabolism in the injured region through of the use of well-known techniques such as the compartmental modeling.
Radiomics in Radiation Oncology
Definition of radiotherapy target volume is a critical step in treatment planning for all tumor sites. Conventional magnetic resonance imaging (MRI) pulse sequences are used for the definition of gross target volume (GTV) and contouring of glioblastoma multiforme (GBM), meningioma and, other types of intracranial tumors. We are using multiparametric MRI combined with radiomic features to improve the texture-based differentiation of tumor from edema for GTV definition, and to differentiate vasogenic from tumor cell infiltration edema.
So far, it has been possible to select a small number of radiological texture characteristics from a set of several hundred parameters initially calculated through different scenarios, different MRI sequences and various approaches. The selected parameters allowed the segregation of the tumor regions in the brain and the differentiation of edema and tumor tissue. Additional regions of interest (e.g. necrotic region, etc.) will be included in our next studies.
In addition, we are performing an assessment using different modalities (MRI, CT, PET) correlated with histopathology, pre and post dosimetry planning and tumor response evaluation.
Dosimetry in Y-90 Radioembolization
Current imaging assessment requires one to two months post-treatment to determine effectiveness due to radical changes that occur to the tissue in response to radiation. This delay is a hindrance to providing better patient outcomes, and, thus, there is a need to identify faster biomarkers to determine treatment success.
We will assess Y-90 dosimetry preplanning using MRI, CT and SPECT MAA images using a software called Simplicity. Then, the result will be correlated with post dosimetry map using PET and tumor response based on physician evaluation and histopathology results.