Author name cluster

Priyanka Bhatt

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

2 author rows

YNICL Journal 2025 Journal Article

Network-targeted transcranial magnetic stimulation (TMS) for mild cognitive impairment (MCI)

Joy L. Taylor
Priyanka Bhatt
Beatriz Hernandez
Michael Iv
Maheen M. Adamson
Alesha Heath
Jerome A. Yesavage
Margaret Windy McNerney

BACKGROUND: Transcranial magnetic stimulation (TMS) is a promising non-pharmacological intervention for treatment of mild cognitive impairment (MCI) and early Alzheimer's disease (AD). Yet, we know little about precisely where stimulation would be ideal to improve cognitive function. OBJECTIVE: To examine the network functional connectivity (fc) characteristics of prefrontal and parietal stimulation sites, given that these sites have led to improved cognitive function in TMS studies involving MCI-AD and unimpaired participants. METHODS: Resting-state functional MRI data were acquired from 32 MCI participants at the baseline visit of an ongoing TMS trial and used to compute connectivity with prefrontal and parietal stimulation locations, selected on the basis of previous TMS studies. The TMS seed maps were examined for extent of spatial overlap with eight canonical networks. After identifying the network most likely to be targeted by TMS, we applied strategies that may provide purer targeting. Finally, we examined network connectivity in relation to participants' behavioral characteristics because of the potential for TMS treatment to be personalized. RESULTS: The prefrontal TMS seed map overlapped primarily with the salience network. The prefrontal site is also notable for its anti-correlated connectivity with the AD-vulnerable posterior cingulate cortex (PCC). The parietal TMS seed map showed the expected strong positive connectivity with the PCC and other default network regions. Nonetheless, this particular parietal site may simultaneously modulate the fronto-parietal network. Strategies to improve network targeting and to personalize TMS are reported as secondary findings. CONCLUSION: These results can be applied to network-targeted brain stimulation for MCI and early AD treatment. Greater precision and personalization of TMS offer the promise of achieving better outcomes for individuals with MCI or mild AD dementia.

Details DOI

ECAI Conference 2023 Conference Paper

Building a Few-Shot Cross-Domain Multilingual NLU Model for Customer Care

Neeraj Agrawal
Priyanka Bhatt

Customer care is an essential pillar of the e-commerce shopping experience with companies spending millions of dollars each year, employing automation and human agents, across geographies (like US, Canada, Mexico, Chile), channels (like Chat, Interactive Voice Response (IVR)), and languages (like English, Spanish). SOTA pre-trained models like multilingual-BERT, fine-tuned on annotated data have shown good performance in downstream tasks relevant to Customer Care. However, model performance is largely subject to the availability of sufficient annotated domain-specific data. Cross-domain availability of data remains a bottleneck, thus building an intent classifier that generalizes across domains (defined by channel, geography, and language) with only a few annotations, is of great practical value. In this paper, we propose an embedder-cum-classifier model architecture which extends state-of-the-art domain-specific models to other domains with only a few labeled samples. We adopt a supervised fine-tuning approach with isotropic regularizers to train a domain-specific sentence embedder and a multilingual knowledge distillation strategy to generalize this embedder across multiple domains. The trained embedder, further augmented with a simple linear classifier can be deployed for new domains. Experiments on Canada and Mexico e-commerce Customer Care dataset with few-shot intent detection show an increase in accuracy by 20-23% against the existing state-of-the-art pre-trained models.

Details

AAAI Conference 2015 Conference Paper

A Stackelberg Game Approach for Incentivizing Participation in Online Educational Forums with Heterogeneous Student Population

Rohith Vallam
Priyanka Bhatt
Debmalya Mandal
Narahari Y.

Increased interest in web-based education has spurred the proliferation of online learning environments. However, these platforms suffer from high dropout rates due to lack of sustained motivation among the students taking the course. In an effort to address this problem, we propose an incentive-based, instructor-driven approach to orchestrate the interactions in online educational forums (OEFs). Our approach takes into account the heterogeneity in skills among the students as well as the limited budget available to the instructor. We first analytically model OEFs in a non-strategic setting using ideas from lumpable continuous time Markov chains and compute expected aggregate transient net-rewards for the instructor and the students. We next consider a strategic setting where we use the rewards computed above to set up a mixed-integer linear program which views an OEF as a single-leader-multiplefollowers Stackelberg game and recommends an optimal plan to the instructor for maximizing student participation. Our experimental results reveal several interesting phenomena including a striking non-monotonicity in the level of participation of students vis-a-vis the instructor’s arrival rate.

PDF Details

YNIMG Journal 2015 Journal Article

An algorithm for optimal fusion of atlases with different labeling protocols

Juan Eugenio Iglesias
Mert Rory Sabuncu
Iman Aganj
Priyanka Bhatt
Christen Casillas
David Salat
Adam Boxer
Bruce Fischl

In this paper we present a novel label fusion algorithm suited for scenarios in which different manual delineation protocols with potentially disparate structures have been used to annotate the training scans (hereafter referred to as “atlases”). Such scenarios arise when atlases have missing structures, when they have been labeled with different levels of detail, or when they have been taken from different heterogeneous databases. The proposed algorithm can be used to automatically label a novel scan with any of the protocols from the training data. Further, it enables us to generate new labels that are not present in any delineation protocol by defining intersections on the underling labels. We first use probabilistic models of label fusion to generalize three popular label fusion techniques to the multi-protocol setting: majority voting, semi-locally weighted voting and STAPLE. Then, we identify some shortcomings of the generalized methods, namely the inability to produce meaningful posterior probabilities for the different labels (majority voting, semi-locally weighted voting) and to exploit the similarities between the atlases (all three methods). Finally, we propose a novel generative label fusion model that can overcome these drawbacks. We use the proposed method to combine four brain MRI datasets labeled with different protocols (with a total of 102 unique labeled structures) to produce segmentations of 148 brain regions. Using cross-validation, we show that the proposed algorithm outperforms the generalizations of majority voting, semi-locally weighted voting and STAPLE (mean Dice score 83%, vs. 77%, 80% and 79%, respectively). We also evaluated the proposed algorithm in an aging study, successfully reproducing some well-known results in cortical and subcortical structures.

Details DOI

YNIMG Journal 2015 Journal Article

Bayesian segmentation of brainstem structures in MRI

Juan Eugenio Iglesias
Koen Van Leemput
Priyanka Bhatt
Christen Casillas
Shubir Dutt
Norbert Schuff
Diana Truran-Sacrey
Adam Boxer

In this paper we present a method to segment four brainstem structures (midbrain, pons, medulla oblongata and superior cerebellar peduncle) from 3D brain MRI scans. The segmentation method relies on a probabilistic atlas of the brainstem and its neighboring brain structures. To build the atlas, we combined a dataset of 39 scans with already existing manual delineations of the whole brainstem and a dataset of 10 scans in which the brainstem structures were manually labeled with a protocol that was specifically designed for this study. The resulting atlas can be used in a Bayesian framework to segment the brainstem structures in novel scans. Thanks to the generative nature of the scheme, the segmentation method is robust to changes in MRI contrast or acquisition hardware. Using cross validation, we show that the algorithm can segment the structures in previously unseen T1 and FLAIR scans with great accuracy (mean error under 1mm) and robustness (no failures in 383 scans including 168 AD cases). We also indirectly evaluate the algorithm with a experiment in which we study the atrophy of the brainstem in aging. The results show that, when used simultaneously, the volumes of the midbrain, pons and medulla are significantly more predictive of age than the volume of the entire brainstem, estimated as their sum. The results also demonstrate that the method can detect atrophy patterns in the brainstem structures that have been previously described in the literature. Finally, we demonstrate that the proposed algorithm is able to detect differential effects of AD on the brainstem structures. The method will be implemented as part of the popular neuroimaging package FreeSurfer.

Details DOI

YNIMG Journal 2010 Journal Article

High-dimensional pattern regression using machine learning: From medical images to continuous clinical variables

Ying Wang
Yong Fan
Priyanka Bhatt
Christos Davatzikos

This paper presents a general methodology for high-dimensional pattern regression on medical images via machine learning techniques. Compared with pattern classification studies, pattern regression considers the problem of estimating continuous rather than categorical variables, and can be more challenging. It is also clinically important, since it can be used to estimate disease stage and predict clinical progression from images. In this work, adaptive regional feature extraction approach is used along with other common feature extraction methods, and feature selection technique is adopted to produce a small number of discriminative features for optimal regression performance. Then the Relevance Vector Machine (RVM) is used to build regression models based on selected features. To get stable regression models from limited training samples, a bagging framework is adopted to build ensemble basis regressors derived from multiple bootstrap training samples, and thus to alleviate the effects of outliers as well as facilitate the optimal model parameter selection. Finally, this regression scheme is tested on simulated data and real data via cross-validation. Experimental results demonstrate that this regression scheme achieves higher estimation accuracy and better generalizing ability than Support Vector Regression (SVR).

Details DOI