The problem of real-time gesture from videos is explored, using techniques of dimension reduction to approach this by using methods analogous to those developed for detecting objects in images and voices in sound data. A variety of innovative pre-processing techniques significantly downsize the total video data. This makes the use of the relatively computationally expensive Higher-Order Singular Value Decomposition (HOSVD) for the classification process feasible. The HOSVD transforms the data into a mathematical space where the classification process becomes much simpler. Once the input video of a gesture is transformed, it is tested against a training set that was created a priori and specially indexed to optimize the search. This process is a refined extension of the ‘eigenfaces’ technique developed for face recognition. The performance is compared to other contemporary methods and its feasibility for larger gesture databases discussed. It is our pre-processing which makes SVD feature selection possible and makes our method robust to variations in translation, rotation, and scale of the gesture in the video. The primary difficulties with scale include both temporal and spatial variations, specifically the rate of the gesture over the video length and the size of the hands performing the gesture. It is shown that our pre-processing steps sufficiently handle the aforementioned challenges except for the variations in time. Comparisons are made between various methods and a balance between computational load and accuracy is discussed. Minimizing the computations is essential for real-time recognition and we intend to release our core code to be incorporated into other computer vision projects.

Parkinson’s disease is a motor disorder characterized by tremors, rigidity, and inability to initiate movement. One therapy for non-drug responsive cases is high amplitude, high frequency stimulation to the basal ganglia’s subthalamic nucleus (STN), via surgically implanted electrodes. While the mechanisms of how this therapy restores proper motor function remains under debate, one hypothesis is that the currents disrupt rhythmic, synchronous firing patterns present in basal ganglia structures of people with Parkinson’s disease. Our goal is to explore the effects of low amplitude, low frequency stimulation to the basal ganglia nucleus globus pallidus externus (Gpe), specifically whether this type of stimulation can disrupt pathological firing synchrony in GPe and STN. We use a conductance-based model of individual neurons in multicellular networks with varying patterns of short, depolarizing currents to mimic deep brain stimulation. Through this work we hope to discover novel patterns of deep-brain stimulation with the potential to eliminate pathological firing patterns of cells in the basal ganglia.

Glioblastoma Multiforme (GBM) is an aggressive type of primary brain tumor, which is treated by surgery followed by chemotherapy and radiation. A majority of cases prove to be highly resistant resulting in uniform fatality. At present, research has been moving towards developing a patient-centered model of treatment that is developed by finding the unique characteristics of a patient's tumor growth kinetics and response to therapy. GBM is a very heterogeneous disease despite having a generally poor prognosis. A promising method of investigating this heterogeneity has been through finding patterns in the molecular-genetic makeup of these tumors. Large-scale mathematical analysis on data from open-source repositories such as The Cancer Genome Atlas (TCGA) has already led to the discovery and validation of four GBM subclasses which point to characteristic areas of origination in the brain as well as differential growth behaviour. A limitation in this strictly data-oriented approach is the lack of meaningful clinical perspective applied which has resulted in the findings only being directed towards gauging patient survival. The focus of our research has been the discovery of information about individual characteristics of patients tumors that may have more clinical significance for each patient. Our hypothesis is that there is a correlation between the different biological subtypes classified by cluster analysis and the different kinetics of tumor growth measured by imaging. This work will be done by developing tools to investigate the genetic changes that may be driving different kinetics of disease progression (the proliferation and migration rates of the tumor cells) measured from actual patient imaging. This approach is guided by the belief that individual patient tumor biology can be used to inform the data analysis techniques and ultimately bring us closer to the goal of identifying the most discriminating treatment to effectively treat an individual's cancer.

Gliomas are invasive, fatal brain tumors that can be detected using magnetic resonance imaging (MRI). However, MRI does not reveal the entire size of tumor. By using a mathematical model developed by Kristin R. Swanson and colleagues, glioma growth can be quantified from the diffusion rate of the tumors cells, the net proliferation rate, and the carrying capacity of the tissue. These parameters are unique to each patient and can be determined by measuring the tumor volume from different MRI modalities as they change over time. Our lab collects data from patients consented to our study and records information about what treatments a given patient receives, since this can affect how the image appears. In my role as a member of the measurement team, I examine patient images to determine the extent of tumor burden on a given date. To obtain these measurements, I use Calcmri, a MATLAB program developed by our lab. Depending on the modality, enhancement on MRI slices can indicate tumor volume as well as other fluids. Calcmri can adjust the intensity threshold at which MRI enhancement will appear, which allows me to isolate the tumor and specify a region for the program to measure. I measure a series of two-dimensional cross section images in order to calculate a combined into 3-dimensional volume. These measurements are used to model the growth of the tumor, “in silico”. Our lab uses information from these models to discover clinical applications, such as to suggest more accurate treatment dosages and to predict life expectancy, which follows an overall goal to better capture the extent and invasion of tumor burden than what appears on MRI.

Gliomas are a uniformly fatal type of primary brain cancer characterized by their diffuse and invasive nature. Dr. Kristin Swanson and her team have developed a mathematical model of glioma growth that uses an individual’s unique tumor characteristics to derive metrics that enable the prediction of patient specific growth patterns and survival times. To date, relatively few studies have looked at predicting clinical progression of disease, which may be a function of disease localization. Clinical progression of disease incorporates variables such as mental cognition and impairment and is typically quantified using the widely used Karnofsky Performance Score (KPS). KPS is an accepted prognostic marker of a patient’s likelihood of tolerating chemotherapy and radiotherapy. In one study accessing functional status after gross total resection (GTR) of a patient’s enhancing tumor region, only 28% of patients had higher KPS post-operatively. 49% had stable scores, while 23% had declines in KPS of 10 to 30 points (Barker 1998). The variability in post GTR KPS scores suggests that there are patient-specific drivers of KPS. We will do a multivariate analysis to investigate which functional brain regions influence the mental and physical impairments that determine KPS scoring. We will identify brain regions, “functional drivers of KPS”, where tumor presence, as calculated by Dr. Swanson’s mathematical model, is well associated with a particular range of KPS scores. For each functional driver, we will use the region’s tumor cell concentration at time of tumor diagnosis, to quantify the future KPS. We will compare the predicted KPS to the actual KPS. If the initial and next time points are located pre and post- radiotherapy, chemotherapy, or resection, this will allow us another way of assessing the efficacy of therapies in palliating symptoms.