Dipesh Pokharel, PhD

Dipesh Pokharel, PhD

orcid
The University of Toledo, College of Medicine

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 March 2025 in AI in Neuroscience
Verified
Journal Article
Closed access
NRNabeel Rangoonwala
Nabeel Rangoonwala
KLKhoi Le
Khoi Le
VPVaibhavi Peshattiwar
Vaibhavi Peshattiwar
CSCaroline Swain
Caroline Swain
Dipesh Pokharel, PhDDipesh Pokharel, PhD
Dipesh Pokharel, PhD
KVKala Venkiteswaran
Kala Venkiteswaran
Background: Behavioral assessment of parkinsonism often relies on human rater evaluation. However, human biases and variability necessitate larger sample sizes to maintain validity, leading to extensive video analysis and limiting researchers’ time. Recent artificial intelligence (AI) and machine learning (ML) advancements enable efficient data analysis, offering unbiased decision-making and consistency across scenarios, bridging inter-rater differences. While not fully automating jobs, AI/ML boosts productivity when properly trained with diverse data. This study aims to show that AI/ML can assist in the analysis of rat parkinsonian behavioral studies to reduce labor dependence while still maintaining accuracy. Methods: DeepLabCut (DLC), an animal pose estimation software, was used to analyze motor behavior in video recordings of parkinsonian Sprague Dawley rats while they performed the stepping test ( n = 24). The stepping test involves observing the animal’s locomotor function and motor coordination while it is guided across a flat surface. The amount of adjusting steps was counted over the 1-meter distance. Twenty-eight videos ( n = 24 + 4 training videos) were fed into DLC, which then selected 20 frames per video using a k-nearest neighbors’ algorithm and subsequently labeled to train the model. This one-time training process took 3 h. The output, which has the tracked coordinates of the forepaw being tested, was fed into a script in R to plot Δy between consecutive frames. The positive peaks were counted as one step, and large negative peaks were counted as a reset or side switch. The counts for each video were then compared with an independent manual rater. Results: There was good absolute agreement between the two scoring methods, using the two-way random effect model, kappa = 0.9, p < 0.0001. It takes 10–15 min to go through each video manually, but the DLC-assisted scoring resulted in 3–4 min per video. These results show that DLC-assisted scoring produced results that could be on par with manual scoring. In addition, this shows a feasible avenue to integrate AI/ML in parkinsonian behavioral studies to reduce the workload for analysis and eventually, fully automating such tasks.

21 March 2026 in NeuroToxicology
Verified
Journal Article
Open access
CSCaroline C. Swain
Caroline C. Swain
SPStephen Prevoznik
Stephen Prevoznik
JKJulia A. Kalbus
Julia A. Kalbus
Dipesh Pokharel, PhDDipesh Pokharel, PhD
Dipesh Pokharel, PhD
DBDilshan Beligala
Dilshan Beligala
TSThyagarajan Subramanian
Thyagarajan Subramanian
The relationship between paraquat exposure and Parkinson's disease (PD) has been extensively studied. Numerous paraquat-induced rodent models of PD have been developed, however, substantial variability in dosing regimens and study design complicates interpretation. Improving the translational relevance of these models is critical. The aim of this systematic review and meta-analysis was to evaluate the face validity of paraquat-induced rodent models of PD. Studies which quantitatively evaluated substantia nigra pars compacta (SNpc) neuronal loss and behavioral motor deficits were included. An electronic search in three databases was performed. A systematic review and meta-analysis was performed of 54 studies (84 experimental groups). Random-effects meta-analyses of continuous data showed a significant effect of SNpc neuronal loss (SMD=3.07, 95% CI 2.48, 3.65; I2=78.1%; p < 0.0001) and motor deficit in studies using the rotarod test, open field test, rearing, and gait analysis, but not the pole test. Subgroup analyses and meta-regression for varying studying characteristics did not account for all heterogeneity. A random-effects meta-analysis of binary data for the event that a significant motor deficit was observed according to the percent SNpc neuron loss revealed a pooled odds ratio (OR=5.08, 95% CI 0.96, 26.83; I2=71.4%; p = 0.0009) and favored a motor deficit in experimental groups that had 26% or greater SNpc neuronal loss. We discuss here how varying study characteristics contribute to the heterogeneity in results across studies. Our findings support the continued use of paraquat-induced rodent models of PD as translationally relevant models, and we highlight the importance of applying appropriate study characteristics in future studies.

1 April 2026 in Neuroscience Letters
Verified
Journal Article
Open access
CSCaroline Swain
Caroline Swain
Dipesh Pokharel, PhDDipesh Pokharel, PhD
Dipesh Pokharel, PhD
VPVaibhavi Peshattiwar
Vaibhavi Peshattiwar
KLKhoi Le
Khoi Le
IKIsabel Kennedy
Isabel Kennedy
TSThyagarajan Subramanian
Thyagarajan Subramanian
Recent studies have identified a "body-first" subtype of Parkinson's disease (PD) that is associated with more rapid progression to Parkinson's disease dementia (PDD). It is currently unknown whether chronic levodopa, the primary treatment in PD, contributes to the progression of disease in the "body-first" subtype of PD. We investigated whether chronic levodopa affects motor or cognitive function in the paraquat and lectin (P + L) rat model of PD, a model which recapitulates the "body-first" subtype of PD. Rats (n = 9) were administered 7 days of P + L (1 mg/kg P + 0.05% L in 1% sucrose solution, oral gavage) with injection of cholecystokinin (3 μg/kg CCK, ip) to induce parkinsonism. At 15 weeks post-P + L treatment, rats with bilateral forelimb motor deficit (n = 6) were administered 60 days of twice daily levodopa (4 mg/kg) + benserazide (15 mg/kg) ip injections. Motor and cognitive behavioral testing were done pre-levodopa, on levodopa, and post-levodopa washout. Rats were euthanized and brains stained for tyrosine hydroxylase (TH), choline acetyltransferase (ChAT) and phosphorylated-S129-α-synuclein (pSyn) to investigate the effects of chronic levodopa on neurodegeneration and pathology in the nigrostriatal pathways and basal forebrain nuclei. We show no long-term effects of chronic levodopa treatment on the progression of parkinsonian motor deficits, visuospatial working memory, dopaminergic nigral degeneration, or cholinergic medial septal nucleus - vertical nucleus of diagonal band of Broca complex degeneration in the P + L rat model of PD. This study provides important preclinical evidence that chronic levodopa therapy does not cause progression of neurodegeneration in this rat model which recapitulates the "body-first" subtype of PD.

28 July 2025 in SSRN Electronic Journal
Verified
Preprint
Open access
Dipesh Pokharel, PhDDipesh Pokharel, PhD
Dipesh Pokharel, PhD
KLKhoi Le
Khoi Le
DBDilshan Beligala
Dilshan Beligala
TSThyagarajan Subramanian
Thyagarajan Subramanian
KVKala Venkiteswaran
Kala Venkiteswaran

10 February 2026 in bioRxiv
Verified
Preprint
Open access
PDF file
Dipesh Pokharel, PhDDipesh Pokharel, PhD
Dipesh Pokharel, PhD
KLKhoi Le
Khoi Le
DBDilshan H. Beligala
Dilshan H. Beligala
TSThyagarajan Subramanian
Thyagarajan Subramanian
KVKala Venkiteswaran
Kala Venkiteswaran
Paw preference, or handedness, is a widely studied behavioral trait used to assess lateralization and motor function in rodents. This study aimed to determine the consistency and reliability of three commonly used behavioral tests to rigorously assess paw preference: the Collins Test, the Staircase Test, and the Pawedness Trait Test. Thirty Sprague Dawley rats (12-48 weeks; 20 females, 10 males) were subjected to all three behavioral tests. Paw uses were recorded, and the laterality index was calculated for each test. Additional cohorts of younger rats (6-9 weeks; 45 females, 45 males) and older rats (12-48 weeks; 38 females, 45 males) were tested to assess the effects of age and sex on paw preference. ANOVA, Fleiss and pairwise Cohen kappa statistics were used for statistical analysis. All three tests yielded comparable measures of paw preference (ANOVA, p = 0.801). Substantial inter-test agreement was demonstrated by Fleiss kappa (k = 0.761, p = 3.93 x 10^-12). Paw preference did not significantly vary by age or sex, and the distribution of left, right, and ambidextrous preference categories aligned with existing literature. The Collins, Staircase, and Pawedness Trait Tests provide consistent and reliable assessments of paw preference in Sprague Dawley rats. These validated behavioral assays can serve as essential tools for preclinical research, including models of motor asymmetry observed in stroke, cerebral palsy, traumatic brain injury, language lateralization, and neurodegenerative disease.

18 November 2024 in Ecological Psychology
Verified
Journal Article
Closed access
TBT. R. Brooks
T. R. Brooks
Dipesh Pokharel, PhDDipesh Pokharel, PhD
Dipesh Pokharel, PhD