LULavanya Umapathy

Lavanya Umapathy

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Lavanya Umapathy has a total of 47 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Automated Coregistered Segmentation for Volumetric Analysis of Multiparametric Renal MRI

    by Aya Ghoul, Cecilia Liang, Isabelle Loster, Lavanya Umapathy, Bernd Kühn, Petros Martirosian, Ferdinand Seith, Sergios Gatidis, Thomas Küstner

    Purpose: This study aims to develop and evaluate a fully automated deep learning‐driven postprocessing pipeline for multiparametric renal MRI, enabling accurate kidney alignment, segmentation, and quantitative feature extraction within a single efficient workflow. Methods: Our method has three main stages. First, a segmentation network delineates renal structures in high‐contrast images. Next, a deep learning‐based pairwise image registration algorithm maps the multiparametric image series to a common target and transfers the predicted annotations between the multiparametric images. Finally, clinically relevant quantitative parameters are extracted through region‐specific assessment of renal structure and function based on the aligned and segmented multiparametric data. We used five‐fold cross‐validation to compare the segmentation outcomes and extracted features with manual analyses in 24 patients with prostate cancer or neuroendocrine tumors and 10 healthy subjects, each undergoing repeated scans. Results: Our automated pipeline achieved high agreement with expert kidney segmentation while delivering significant alignment improvements through registration. Volumetric analysis showed a strong correlation ( r 0.9) with manual results, and feature extraction demonstrated high intraclass correlation coefficients with minimal bias. The complete processing pipeline, encompassing coregistration, segmentation, and feature extraction, required approximately 15 s per scan from raw input to final quantitative output. Conclusion: The study establishes a reliable automated pipeline for renal multiparametric MRI postprocessing. The achieved accuracy and efficiency can support improved diagnosis and treatment planning for patients with kidney disease.


  • Artificial Intelligence in Prostate MRI : Addressing Current Limitations Through Emerging Technologies

    by Patricia M. Johnson, Lavanya Umapathy, Bradley Gigax, Juan Kochen Rossi, Angela Tong, Mary Bruno, Daniel K. Sodickson, Madhur Nayan, Hersh Chandarana

    Prostate MRI has transformed lesion detection and risk stratification in prostate cancer, but its impact is constrained by the high cost of the exam, variability in interpretation, and limited scalability. False negatives, false positives, and moderate inter‐reader agreement undermine reliability, while long acquisition times restrict throughput. Artificial intelligence (AI) offers potential solutions to address many of the limitations of prostate MRI in the clinical management pathway. Machine learning–based triage can refine patient selection to optimize resources. Deep learning reconstruction enables accelerated acquisition while preserving diagnostic quality, with multiple FDA‐cleared products now in clinical use. Ongoing development of automated quality assessment and artifact correction aims to improve reliability by reducing nondiagnostic exams. In image interpretation, AI models for lesion detection and clinically significant prostate cancer prediction achieve performance comparable to radiologists, and the PI‐CAI international reader study has provided the strongest evidence to date of non‐inferiority at scale. More recent work extends MRI‐derived features into prognostic modeling of recurrence, metastasis, and functional outcomes. This review synthesizes progress across five domains—triage, accelerated acquisition and reconstruction, image quality assurance, diagnosis, and prognosis—highlighting the level of evidence, validation status, and barriers to adoption. While acquisition and reconstruction are furthest along, with FDA‐cleared tools and prospective evaluations, triage, quality control, and prognosis remain earlier in development. Ensuring equitable performance across populations, incorporating uncertainty estimation, and conducting prospective workflow trials will be essential to move from promising prototypes to routine practice. Ultimately, AI could accelerate the adoption of prostate MRI toward a scalable platform for earlier detection and population‐level prostate cancer management. Evidence Level: N/A Technical Efficacy: 3


  • Leveraging Representation Learning for Bi-parametric Prostate MRI to Disambiguate PI-RADS 3 and Improve Biopsy Decision Strategies

    by Lavanya Umapathy, Patricia M. Johnson, Tarun Mtech Dutt, Angela Tong, Sumit Chopra, Daniel K. Sodickson, Hersh Chandarana

    Objectives: Despite its high negative predictive value (NPV) for clinically significant prostate cancer (csPCa), MRI suffers from a substantial number of false positives, especially for intermediate-risk cases. In this work, we determine whether a deep learning model trained with PI-RADS–guided representation learning can disambiguate the PI-RADS 3 classification, detect csPCa from bi-parametric prostate MR images, and avoid unnecessary benign biopsies. Materials and Methods: This study included 28,263 MR examinations and radiology reports from 21,938 men imaged for known or suspected prostate cancer between 2015 and 2023 at our institution (21 imaging locations with 34 readers), with 6352 subsequent biopsies. We trained a deep learning model, a representation learner (RL), to learn how radiologists interpret conventionally acquired T2-weighted and diffusion-weighted MR images, using exams in which the radiologists are confident in their risk assessments (PI-RADS 1 and 2 for the absence of csPCa vs. PI-RADS 4 and 5 for the presence of csPCa, n=21,465). We then trained biopsy-decision models to detect csPCa (Gleason score ≥7) using these learned image representations, and compared them to the performance of radiologists, and of models trained on other clinical variables (age, prostate volume, PSA, and PSA density) for treatment-naïve test cohorts consisting of only PI-RADS 3 (n=253, csPCa=103) and all PI-RADS (n=531, csPCa=300) cases. Results: On the 2 test cohorts (PI-RADS-3-only, all-PI-RADS), RL-based biopsy-decision models consistently yielded higher AUCs in detecting csPCa (AUC=0.73 [0.66, 0.79], 0.88 [0.85, 0.91]) compared with radiologists (equivocal, AUC=0.79 [0.75, 0.83]) and the clinical model (AUCs=0.69 [0.62, 0.75], 0.78 [0.74, 0.82]). In the PIRADS-3-only cohort, all of whom would be biopsied using our institution’s standard of care, the RL decision model avoided 41% (62/150) of benign biopsies compared with the clinical model (26%, P P <0.001) with comparable sensitivity (93% vs. 92%), higher NPV (0.87 vs. 0.77), and biopsy yield (0.75 vs. 0.64). The combination of clinical and RL decision models further avoided benign biopsies (46% in PI-RADS-3-only and 62% in all-PI-RADS) while improving NPV (0.82, 0.88) and biopsy yields (0.52, 0.76) across the 2 test cohorts. Conclusions: Our PI-RADS–guided deep learning RL model learns summary representations from bi-parametric prostate MR images that can provide additional information to disambiguate intermediate-risk PI-RADS 3 assessments. The resulting RL-based biopsy decision models also outperformed radiologists in avoiding benign biopsies while maintaining comparable sensitivity to csPCa for the all-PI-RADS cohort. Such AI models can easily be integrated into clinical practice to supplement radiologists’ reads in general and improve biopsy yield for any equivocal decisions.


  • Prostate Cancer Risk Stratification and Scan Tailoring Using Deep Learning on Abbreviated Prostate MRI

    by Patricia M. Johnson, Tarun Dutt, Luke A. Ginocchio, Amanpreet Singh Saimbhi, Lavanya Umapathy, Kai Tobias Block, Daniel K. Sodickson, Sumit Chopra, Angela Tong, Hersh Chandarana

    Background: MRI plays a critical role in prostate cancer (PCa) detection and management. Bi‐parametric MRI (bpMRI) offers a faster, contrast‐free alternative to multi‐parametric MRI (mpMRI). Routine use of mpMRI for all patients may not be necessary, and a tailored imaging approach (bpMRI or mpMRI) based on individual risk might optimize resource utilization. Purpose:To develop and evaluate a deep learning (DL) model for classifying clinically significant PCa (csPCa) using bpMRI and to assess its potential for optimizing MRI protocol selection by recommending the additional sequences of mpMRI only when beneficial.Study Type: Retrospective and prospective.Population: The DL model was trained and validated on 26,129 prostate MRI studies. A retrospective cohort of 151 patients (mean age 65 ± 8) with ground‐truth verification from biopsy, prostatectomy, or long‐term follow‐up, alongside a prospective cohort of 142 treatment‐naïve patients (mean age 65 ± 9) undergoing bpMRI, was evaluated.Field Strength/Sequence: 3 T, Turbo‐spin echo T2‐weighted imaging (T2WI) and single shot EPI diffusion‐weighted imaging (DWI).Assessment: The DL model, based on a 3D ResNet‐50 architecture, classified csPCa using PI‐RADS ≥ 3 and Gleason ≥ 7 as outcome measures. The model was evaluated on a prospective cohort labeled by consensus of three radiologists and a retrospective cohort with ground truth verification based on biopsy or long‐term follow‐up. Real‐time inference was tested on an automated MRI workflow, providing classification results directly at the scanner.Statistical Tests: AUROC with 95% confidence intervals (CI) was used to evaluate model performance.Results: In the prospective cohort, the model achieved an AUC of 0.83 (95% CI: 0.77–0.89) for PI‐RADS ≥ 3 classification, with 93% sensitivity and 54% specificity. In the retrospective cohort, the model achieved an AUC of 0.86 (95% CI: 0.80–0.91) for Gleason ≥ 7 classification, with 93% sensitivity and 62% specificity. Real‐time implementation demonstrated a processing latency of 14–16 s for protocol recommendations.Data Conclusion: The proposed DL model identifies csPCa using bpMRI and integrates it into clinical workflows.Evidence Level: 1.Technical Efficacy: Stage 2.


  • Reducing annotation burden in MR: A novel MR‐contrast guided contrastive learning approach for image segmentation

    by Lavanya Umapathy, Taylor Brown, Raza Mushtaq, Mark Greenhill, J'RIck Lu, Diego Martin, Maria Altbach, Ali Bilgin

    Background: Contrastive learning, a successful form of representational learning, has shown promising results in pretraining deep learning (DL) models for downstream tasks. When working with limited annotation data, as in medical image segmentation tasks, learning domain‐specific local representations can further improve the performance of DL models.Purpose: In this work, we extend the contrastive learning framework to utilize domain‐specific contrast information from unlabeled Magnetic Resonance (MR) images to improve the performance of downstream MR image segmentation tasks in the presence of limited labeled data.Methods: The contrast in MR images is controlled by underlying tissue properties (e.g., T1 or T2) and image acquisition parameters. We hypothesize that learning to discriminate local representations based on underlying tissue properties should improve subsequent segmentation tasks on MR images. We propose a novel constrained contrastive learning (CCL) strategy that uses tissue‐specific information via a constraint map to define positive and negative local neighborhoods for contrastive learning, embedding this information in the representational space during pretraining. For a given MR contrast image, the proposed strategy uses local signal characteristics (constraint map) across a set of related multi‐contrast MR images as a surrogate for underlying tissue information. We demonstrate the utility of the approach for downstream: (1) multi‐organ segmentation tasks in T2‐weighted images where a DL model learns T2 information with constraint maps from a set of 2D multi‐echo T2‐weighted images (n = 101) and (2) tumor segmentation tasks in multi‐parametric images from the public brain tumor segmentation (BraTS) (n = 80) dataset where DL models learn T1 and T2 information from multi‐parametric BraTS images. Performance is evaluated on downstream multi‐label segmentation tasks with limited data in (1) T2‐weighted images of the abdomen from an in‐house Radial‐T2 (Train/Test = 30/20), (2) public Cartesian‐T2 (Train/Test = 6/12) dataset, and (3) multi‐parametric MR images from the public brain tumor segmentation dataset (BraTS) (Train/Test = 40/50). The performance of the proposed CCL strategy is compared to state‐of‐the‐art self‐supervised contrastive learning techniques. In each task, a model is also trained using all available labeled data for supervised baseline performance.Results: The proposed CCL strategy consistently yielded improved Dice scores, Precision, and Recall metrics, and reduced HD95 values across all segmentation tasks. We also observed performance comparable to the baseline with reduced annotation effort. The t‐SNE visualization of features for T2‐weighted images demonstrates its ability to embed T2 information in the representational space. On the BraTS dataset, we also observed that using an appropriate multi‐contrast space to learn T1+T2, T1, or T2 information during pretraining further improved the performance of tumor segmentation tasks.Conclusions: Learning to embed tissue‐specific information that controls MR image contrast with the proposed constrained contrastive learning improved the performance of DL models on subsequent segmentation tasks compared to conventional self‐supervised contrastive learning techniques. The use of such domain‐specific local representations could help understand, improve performance, and mitigate the scarcity of labeled data in MR image segmentation tasks.


  • A Stacked Generalization of 3D Orthogonal Deep Learning Convolutional Neural Networks for Improved Detection of White Matter Hyperintensities in 3D FLAIR Images

    by L. Umapathy, G.G. Perez-Carrillo, M.B. Keerthivasan, J.A. Rosado-Toro, M.I. Altbach, B. Winegar, C. Weinkauf, A. Bilgin, For The Alzheimer’s Disease Neuroimaging Initiative

    BACKGROUND AND PURPOSE: Accurate and reliable detection of white matter hyperintensities and their volume quantification can provide valuable clinical information to assess neurologic disease progression. In this work, a stacked generalization ensemble of orthogonal 3D convolutional neural networks, StackGen-Net, is explored for improving automated detection of white matter hyperintensities in 3D T2-FLAIR images. MATERIALS AND METHODS: Individual convolutional neural networks in StackGen-Net were trained on 2.5D patches from orthogonal reformatting of 3D-FLAIR (n = 21) to yield white matter hyperintensity posteriors. A meta convolutional neural network was trained to learn the functional mapping from orthogonal white matter hyperintensity posteriors to the final white matter hyperintensity prediction. The impact of training data and architecture choices on white matter hyperintensity segmentation performance was systematically evaluated on a test cohort (n = 9). The segmentation performance of StackGen-Net was compared with state-of-the-art convolutional neural network techniques on an independent test cohort from the Alzheimer’s Disease Neuroimaging Initiative-3 (n = 20). RESULTS: StackGen-Net outperformed individual convolutional neural networks in the ensemble and their combination using averaging or majority voting. In a comparison with state-of-the-art white matter hyperintensity segmentation techniques, StackGen-Net achieved a significantly higher Dice score (0.76 [SD, 0.08], F1-lesion (0.74 [SD, 0.13]), and area under precision-recall curve (0.84 [SD, 0.09]), and the lowest absolute volume difference (13.3% [SD, 9.1%]). StackGen-Net performance in Dice scores (median = 0.74) did not significantly differ (P = .22) from interobserver (median = 0.73) variability between 2 experienced neuroradiologists. We found no significant difference (P = .15) in white matter hyperintensity lesion volumes from StackGen-Net predictions and ground truth annotations. CONCLUSIONS: A stacked generalization of convolutional neural networks, utilizing multiplanar lesion information using 2.5D spatial context, greatly improved the segmentation performance of StackGen-Net compared with traditional ensemble techniques and some state-of-the-art deep learning models for 3D-FLAIR.


  • Fully Automated Segmentation of Globes for Volume Quantification in CT Images of Orbits using Deep Learning

    by L. Umapathy, B. Winegar, L. MacKinnon, M. Hill, M.I. Altbach, J.M. Miller, A. Bilgin

    BACKGROUND AND PURPOSE: Fast and accurate quantification of globe volumes in the event of an ocular trauma can provide clinicians with valuable diagnostic information. In this work, an automated workflow using a deep learning-based convolutional neural network is proposed for prediction of globe contours and their subsequent volume quantification in CT images of the orbits. MATERIALS AND METHODS: An automated workflow using a deep learning -based convolutional neural network is proposed for prediction of globe contours in CT images of the orbits. The network, 2D Modified Residual UNET (MRes-UNET2D), was trained on axial CT images from 80 subjects with no imaging or clinical findings of globe injuries. The predicted globe contours and volume estimates were compared with manual annotations by experienced observers on 2 different test cohorts. RESULTS: On the first test cohort (n = 18), the average Dice, precision, and recall scores were 0.95, 96%, and 95%, respectively. The average 95% Hausdorff distance was only 1.5 mm, with a 5.3% error in globe volume estimates. No statistically significant differences (P = .72) were observed in the median globe volume estimates from our model and the ground truth. On the second test cohort (n = 9) in which a neuroradiologist and 2 residents independently marked the globe contours, MRes-UNET2D (Dice = 0.95) approached human interobserver variability (Dice = 0.94). We also demonstrated the utility of inter-globe volume difference as a quantitative marker for trauma in 3 subjects with known globe injuries. CONCLUSIONS: We showed that with fast prediction times, we can reliably detect and quantify globe volumes in CT images of the orbits across a variety of acquisition parameters.


  • Open Access

    Auditory Processing Deficits Are Selectively Associated with Medial Temporal Lobe Mnemonic Function and White Matter Integrity in Aging Macaques

    by Daniel T Gray, Lavanya Umapathy, Nicole M De La Peña, Sara N Burke, James R Engle, Theodore P Trouard, Carol A Barnes

    Deficits in auditory function and cognition are hallmarks of normative aging. Recent evidence suggests that hearing-impaired individuals have greater risks of developing cognitive impairment and dementia compared to people with intact auditory function, although the neurobiological bases underlying these associations are poorly understood. Here, a colony of aging macaques completed a battery of behavioral tests designed to probe frontal and temporal lobe-dependent cognition. Auditory brainstem responses (ABRs) and visual evoked potentials were measured to assess auditory and visual system function. Structural and diffusion magnetic resonance imaging were then performed to evaluate the microstructural condition of multiple white matter tracts associated with cognition. Animals showing higher cognitive function had significantly better auditory processing capacities, and these associations were selectively observed with tasks that primarily depend on temporal lobe brain structures. Tractography analyses revealed that the fractional anisotropy (FA) of the fimbria-fornix and hippocampal commissure were associated with temporal lobe-dependent visual discrimination performance and auditory sensory function. Conversely, FA of frontal cortex-associated white matter was not associated with auditory processing. Visual sensory function was not associated with frontal or temporal lobe FA, nor with behavior. This study demonstrates significant and selective relationships between ABRs, white matter connectivity, and higher-order cognitive ability.