Browsing by Author "Barua, Souptik"
Now showing 1 - 9 of 9
Results Per Page
Sort Options
Item Baseline leptin predicts response to metformin in adolescents with type 1 diabetes and increased body mass index(Wiley, 2023) Ismail, Heba M.; Barua, Souptik; Wang, Johnny; Sabharwal, Ashutosh; Libman, Ingrid; Bacha, Fida; Nadeau, Kristen J.; Tosur, Mustafa; Redondo, Maria J.Item Computed Tomography Radiomics Kinetics as Early Imaging Correlates of Osteoradionecrosis in Oropharyngeal Cancer Patients(Frontiers Media S.A., 2021) Barua, Souptik; Elhalawani, Hesham; Volpe, Stefania; Al Feghali, Karine A.; Yang, Pei; Ng, Sweet Ping; Elgohari, Baher; Granberry, Robin C.; Mackin, Dennis S.; Gunn, G. Brandon; Hutcheson, Katherine A.; Chambers, Mark S.; Court, Laurence E.; Mohamed, Abdallah S.R.; Fuller, Clifton D.; Lai, Stephen Y.; Rao, ArvindOsteoradionecrosis (ORN) is a major side-effect of radiation therapy in oropharyngeal cancer (OPC) patients. In this study, we demonstrate that early prediction of ORN is possible by analyzing the temporal evolution of mandibular subvolumes receiving radiation. For our analysis, we use computed tomography (CT) scans from 21 OPC patients treated with Intensity Modulated Radiation Therapy (IMRT) with subsequent radiographically-proven ≥ grade II ORN, at three different time points: pre-IMRT, 2-months, and 6-months post-IMRT. For each patient, radiomic features were extracted from a mandibular subvolume that developed ORN and a control subvolume that received the same dose but did not develop ORN. We used a Multivariate Functional Principal Component Analysis (MFPCA) approach to characterize the temporal trajectories of these features. The proposed MFPCA model performs the best at classifying ORN vs. Control subvolumes with an area under curve (AUC) = 0.74 [95% confidence interval (C.I.): 0.61–0.90], significantly outperforming existing approaches such as a pre-IMRT features model or a delta model based on changes at intermediate time points, i.e., at 2- and 6-month follow-up. This suggests that temporal trajectories of radiomics features derived from sequential pre- and post-RT CT scans can provide markers that are correlates of RT-induced mandibular injury, and consequently aid in earlier management of ORN.Item Dysglycemia in adults at risk for or living with non-insulin treated type 2 diabetes: Insights from continuous glucose monitoring(Elsevier, 2021) Barua, Souptik; Sabharwal, Ashutosh; Glantz, Namino; Conneely, Casey; Larez, Arianna; Bevier, Wendy; Kerr, DavidBackground: Continuous glucose monitoring (CGM) has demonstrable benefits for people living with diabetes, but the supporting evidence is almost exclusively from White individuals with type 1 diabetes. Here, we have quantified CGM profiles in Hispanic/Latino adults with or at-risk of non-insulin treated type 2 diabetes (T2D). Methods: 100 participants (79 female, 86% Hispanic/Latino [predominantly Mexican], age 54·6 [±12·0] years) stratified into (i) at risk of T2D, (ii) with pre-diabetes (pre-T2D), and (iii) with non-insulin treated T2D, wore blinded CGMs for 2 weeks. Beyond standardized CGM measures (average glucose, glucose variability, time in 70–140 mg/dL and 70–180 mg/dL ranges), we also examined additional CGM measures based on the time of day. Findings: Standardized CGM measures were significantly different for participants with T2D compared to at-risk and pre-T2D participants (p<0·0001). In addition, pre-T2D participants spent more time between 140 and 180 mg/dL during the day than at-risk participants (p<0·01). T2D participants spent more time between 140 and 180 mg/dL both during the day and overnight compared to at-risk and pre-T2D participants (both p<0·0001). Time in 70–140 mg/dL range during the day was significantly correlated with HbA1c (r=-0·72, p<0·0001), after adjusting for age, sex, BMI, and waist circumference (p<0·0001). Interpretation: Standardized CGM measures show a progression of dysglycemia from at-risk of T2D, to pre-T2D, and to T2D. Stratifying CGM readings by time of day and the range 140–180 mg/dL provides additional metrics to differentiate between the groups. Funding US Department of Agriculture (Grant #2018-33800-28404) and NSF PATHS-UP ERC (Award #1648451).Item Farming for life: impact of medical prescriptions for fresh vegetables on cardiometabolic health for adults with or at risk of type 2 diabetes in a predominantly Mexican-American population(BMJ, 2020) Kerr, David; Barua, Souptik; Glantz, Namino; Conneely, Casey; Kujan, Mary; Bevier, Wendy; Larez, Arianna; Sabharwal, AshutoshIntroduction: Poor diet is the leading cause of poor health in USA, with fresh vegetable consumption below recommended levels. We aimed to assess the impact of medical prescriptions for fresh (defined as picked within 72 hours) vegetables, at no cost to participants on cardiometabolic outcomes among adults (predominantly Mexican-American women) with or at risk of type 2 diabetes (T2D). Methods: Between February 2019 and March 2020, 159 participants (122 female, 75% of Mexican heritage, 31% with non-insulin treated T2D, age 52.5 (13.2) years) were recruited using community outreach materials in English and Spanish, and received prescriptions for 21 servings/week of fresh vegetable for 10 weeks. Pre-post comparisons were made of weight; waist circumference; blood pressure; Hemoglobin A1c (HbA1c, a measure of long-term blood glucose control); self-reported sleep, mood and pain; vegetable, tortilla and soda consumption. After obtaining devices for this study, 66 of 72 participants asked, agreed to wear blinded continuous glucose monitors (CGM). Results: Paired data were available for 131 participants. Over 3 months, waist circumference fell (−0.77 (95% CI −1.42 to 0.12) cm, p=0.022), as did systolic blood pressure (SBP) (−2.42 (95% CI −4.56 to 0.28) mm Hg, p=0.037), which was greater among individuals with baseline SBP >130 mm Hg (−7.5 (95% CI −12.4 to 2.6) mm Hg, p=0.005). Weight reduced by −0.4 (−0.7 to –0.04) kg, p=0.029 among women. For participants with baseline HbA1c >7.0%, HbA1c fell by −0.35 (-0.8 to –0.1), p=0.009. For participants with paired CGM data (n=40), time in range 70–180 mg/dL improved (from 97.4% to 98.9%, p<0.01). Food insecurity (p<0.001), tortilla (p<0.0001) and soda (p=0.013) consumption significantly decreased. Self-reported sleep, mood and pain level scores also improved (all p<0.01). Conclusions: Medical prescriptions for fresh vegetables were associated with clinically relevant improvements in cardiovascular risk factors and quality of life variables (sleep, mood and pain level) in adults (predominantly Mexican-American and female) with or at risk of T2D.Item Hypoglycemia risk with physical activity in type 1 diabetes: a data-driven approach(Frontiers Media S.A., 2023) Prasanna, Sahana; Barua, Souptik; Siller, Alejandro F.; Johnson, Jeremiah J.; Sabharwal, Ashutosh; DeSalvo, Daniel J.Physical activity (PA) provides numerous health benefits for individuals with type 1 diabetes (T1D). However, the threat of exercise-induced hypoglycemia may impede the desire for regular PA. Therefore, we aimed to study the association between three common types of PA (walking, running, and cycling) and hypoglycemia risk in 50 individuals with T1D. Real-world data, including PA duration and intensity, continuous glucose monitor (CGM) values, and insulin doses, were available from the Tidepool Big Data Donation Project. Participants' mean (SD) age was 38.0 (13.1) years with a mean (SD) diabetes duration of 21.4 (12.9) years and an average of 26.2 weeks of CGM data available. We developed a linear regression model for each of the three PA types to predict the average glucose deviation from 70 mg/dl for the 2 h after the start of PA. This is essentially a measure of hypoglycemia risk, for which we used the following predictors: PA duration (mins) and intensity (calories burned), 2-hour pre-exercise area under the glucose curve (adjusted AUC), the glucose value at the beginning of PA, and total bolus insulin (units) within 2 h before PA. Our models indicated that glucose value at the start of exercise and pre-exercise glucose adjusted AUC (p < 0.001 for all three activities) were the most significant predictors of hypoglycemia. In addition, the duration and intensity of PA and 2-hour bolus insulin were weakly associated with hypoglycemia for walking, running, and cycling. These findings may provide individuals with T1D with a data-driven approach to preparing for PA that minimizes hypoglycemia risk.Item Leveraging structure in cancer imaging data using data-driven frameworks to predict clinical outcomes(2019-04-17) Barua, Souptik; Rao, Arvind; Veeraraghavan, AshokImmunotherapy and radiation therapy are two of the most prominent strategies used to treat cancer. While both these strategies have succeeded in treating the disease in many patients and cancer types, they are known to not work well for all patients, sometimes even leading to adverse side effects. There is thus a critical need to be able to predict how patients might respond to these therapies and accordingly design optimal treatment plans. In this thesis, I have built data-driven frameworks that leverage different types of structure in cancer imaging data to predict clinical outcomes of interest. I demonstrate my findings using two kinds of cancer image data: multipexed Immuno-Fluorescence (mIF) images from the field of pathology, and Computed Tomography (CT) from radiology. In mIF images, I quantify spatial structure in terms of proximities of cancer cells and various immune cells by developing ideas from spatial statistics to estimate the nearest neighbor distribution function of a spatial point process. I demonstrate that this quantity, called the G-function, can be used as a visual signature of the infiltration patterns of different cell types of interest. I compute summary metrics from the G-function which are prognostic of clinical outcome in multiple cancer types such as pancreatic, breast, and skin cancer. I design a functional analysis pipeline to more efficiently summarize the G-function and predict the risk of progression in pancreatic cysts. Finally, I propose approaches to capture the heterogeneity in the tumor microenvironment and demonstrate the importance of quantifying spatial structure separately in different tumor regions through a case study in lung cancer. In CT images acquired at different time points, I capture the temporal evolution of a specific class of image features called radiomic features using ideas from functional analysis. I then show that the temporal dynamics of radiomic features can be used to predict clinical outcomes such as the likelihood of complete response to radiation therapy and the risk of developing long-term radiation injuries such as osteoradionecrosis. Overall, I envisage these data-driven frameworks can be incorporated into a clinical setting not only as a tool for prognosis and treatment design but also in basic science research to understand the biological underpinnings of cancer development.Item Rethinking Image Compression for the Object Detection Task(2015-12-03) Barua, Souptik; Veeraraghavan, Ashok; Baraniuk, Richard; Shrivastava, AnshumaliTraditionally, image compression algorithms, such as JPEG, have been designed for human viewers' satisfaction. Increasingly however, more and more images are being viewed by computers, for performing computer vision tasks such as object detection. Image compression and object detection have largely been independent areas of research so far. However, several applications such as surveillance and medical imaging impose severe bandwidth and power restrictions. These constraints make the quality and/or size of the compressed image a critical factor in object detection performance. My works presents three compressed image representations that enable fast and accurate object detection. The first representation is a saliency guided wavelet representation which modifies traditional wavelet compression using the knowledge of saliency to improve both compression and detection performance compared to JPEG images. The second representation, called event stream representation, comes directly from the new DVS sensor which has ultra-low bandwidth and power requirements. We show, for the first time, high speed video reconstruction, and direct detection, on the event data. We achieve detection performance comparable to that on conventional JPEG images. Finally, we explore an abstract compressed representation called patch-wise binary representation, which represents an image (patch-wise) as a collection of short binary strings. We demonstrate two ways of generating these binary strings, called hashing and feature binarization, which enable 10x faster detection. We show promising detection and reconstruction results for both these approaches.Item Temporal changes in bio-behavioral and glycemic outcomes following a produce prescription program among predominantly Hispanic/Latino adults with or at risk of type 2 diabetes(Elsevier, 2023) Sato Imuro, Sandra Emi; Sabharwal, Ashutosh; Conneely, Casey; Glantz, Namino; Bevier, Wendy; Barua, Souptik; Pai, Amruta; Larez, Arianna; Kerr, DavidIn the United States (U.S.), consumption of fresh vegetables and fruits is below recommended levels. Enhancing access to nutritious food through food prescriptions has been recognized as a promising approach to combat diet-related illnesses. However, the effectiveness of this strategy at a large scale remains untested, particularly in marginalized communities where food insecurity rates and the prevalence of health conditions such as type 2 diabetes (T2D) are higher compared to the background population. This study evaluated the impact of a produce prescription program for predominantly Hispanic/Latino adults living with or at risk of T2D. A total of 303 participants enrolled in a 3-month observational cohort received 21 medically prescribed portions/week of fresh produce. A subgroup of 189 participants used continuous glucose monitoring (CGM) to assess the relationship between CGM profile changes and HbA1c level changes. For 247 participants completing the study (76% female, 84% Hispanic/Latino, 32% with T2D, age 56·6 ± 11·9 years), there was a reduction in weight (−1·1 [-1·6 to −0·6] lbs., p < 0.001), waist circumference (−0·4 [-1·0 to 0·6] cm, p = 0·007) and systolic blood pressure (SBP) for participants with baseline SBP >120 mmHg (−4·2 [-6·8 to −1·8] mmHg, p = 0·001). For participants with an HbA1c ≥ 7·0% at baseline, HbA1c fell significantly (−0·5 [-0·9 to −0·1] %, p = 0·01). There were also improvements in food security (p < 0·0001), self-reported ratings of sleep, mood, pain (all p < 0·001), and measures of depression (p < 0·0001), anxiety (p = 0·045), and stress (p = 0·002) (DASS-21). There was significant correlation (r = 0·8, p = 0·001) between HbA1c change and the change in average glucose for participants with worsening HbA1c, but not for participants with an improvement in HbA1c. In conclusion, medical prescription of fresh produce is associated with significant improvements in cardio-metabolic and psycho-social risk factors for Hispanic/Latino adults with or at risk of T2D.Item Towards the characterization of the tumor microenvironment through dictionary learning-based interpretable classification of multiplexed immunofluorescence images(IOP Publishing, 2022) Krishnan, Santhoshi N.; Barua, Souptik; Frankel, Timothy L.; Rao, ArvindObjective. Histology image analysis is a crucial diagnostic step in staging and treatment planning, especially for cancerous lesions. With the increasing adoption of computational methods for image analysis, significant strides are being made to improve the performance metrics of image segmentation and classification frameworks. However, many developed frameworks effectively function as black boxes, granting minimal context to the decision-making process. Thus, there is a need to develop methods that offer reasonable discriminatory power and a biologically-informed intuition to the decision-making process. Approach. In this study, we utilized and modified a discriminative feature-based dictionary learning (DFDL) paradigm to generate a classification framework that allows for discrimination between two distinct clinical histologies. This framework allows us (i) to discriminate between 2 clinically distinct diseases or histologies and (ii) provides interpretable group-specific representative dictionary image patches, or ‘atoms’, generated during classifier training. This implementation is performed on multiplexed immunofluorescence images from two separate patient cohorts- a pancreatic cohort consisting of cancerous and non-cancerous tissues and a metastatic non-small cell lung cancer (mNSCLC) cohort of responders and non-responders to an immunotherapeutic treatment regimen. The analysis was done at both the image-level and subject-level. Five cell types were selected, namely, epithelial cells, cytotoxic lymphocytes, antigen presenting cells, HelperT cells, and T-regulatory cells, as our phenotypes of interest. Results. We showed that DFDL had significant discriminant capabilities for both the pancreatic pathologies cohort (subject-level AUC-0.8878) and the mNSCLC immunotherapy response cohort (subject-level AUC-0.7221). The secondary analysis also showed that more than 50% of the obtained dictionary atoms from the classifier contained biologically relevant information. Significance. Our method shows that the generated dictionary features can help distinguish patients presenting two different histologies with strong sensitivity and specificity metrics. These features allow for an additional layer of model interpretability, a highly desirable element in clinical applications for identifying novel biological phenomena.