Siddharth Madapoosi

Siddharth Madapoosi

Auto-generated profile, based on publicly available information

Siddharth Madapoosi

From the Department of Internal Medicine, Michigan Medicine, Ann Arbor, MI (SM, EW); Department of Family Medicine, Michigan Medicine, Ann Arbor, MI (JGC, AME); ACCESS Madison Family Health Center, Chicago, IL (NI); Division of Allergy and Clinical Immunology, Department of Internal Medicine, Michigan Medicine, Ann Arbor MI (RR)

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
26 September 2024 in Heliyon
Verified
Journal Article
Open access
ATAdam Tisch
Adam Tisch
SMSiddharth Madapoosi
Siddharth Madapoosi
SBStephen Blough
Stephen Blough
JRJan Rosa
Jan Rosa
SESean Eddy
Sean Eddy
FAFadhl Alakwaa
Fadhl Alakwaa
Single-cell RNA sequencing (scRNA-seq) and single-nucleus RNA sequencing (snRNA-seq) provide valuable insights into the cellular states of kidney cells. However, the annotation of cell types often requires extensive domain expertise and time-consuming manual curation, limiting scalability and generalizability. To facilitate this process, we tested the performance of five supervised classification methods for automatic cell type annotation. We analyzed publicly available sc/snRNA-seq datasets from five expert-annotated studies, comprising 62,120 cells from 79 kidney biopsy samples. Datasets were integrated by harmonizing cell type annotations across studies. Five different supervised machine learning algorithms (support vector machines, random forests, multilayer perceptrons, k-nearest neighbors, and extreme gradient boosting) were applied to automatically annotate cell types using four training datasets and one testing dataset. Performance metrics, including accuracy (F1 score) and rejection rates, were evaluated. All five machine learning algorithms demonstrated high accuracies, with a median F1 score of 0.94 and a median rejection rate of 1.8 %. The algorithms performed equally well across different datasets and successfully rejected cell types that were not present in the training data. However, F1 scores were lower when models trained primarily on scRNA-seq data were tested on snRNA-seq data. Despite limitations including the number of biopsy samples, our findings demonstrate that machine learning algorithms can accurately annotate a wide range of adult kidney cell types in scRNA-seq/snRNA-seq data. This approach has the potential to standardize cell type annotation and facilitate further research on cellular mechanisms underlying kidney disease.

1 November 2024 in The Journal of the American Board of Family Medicine
Verified
Journal Article
Open access
PDF file
SMSiddharth Madapoosi
Siddharth Madapoosi
JGJulio Garcia-Castro
Julio Garcia-Castro
NINadia Ilhardt
Nadia Ilhardt
MLMikel Llanes
Mikel Llanes
RRRajan Ravikumar
Rajan Ravikumar
EWEric Walford
Eric Walford
Purpose: Nearly 10% of the United States population has a reported penicillin allergy. However, many of these patients do not have true IgE-mediated allergy and are exposed to alternative antibiotics with increased risks of adverse outcomes, highlighting the importance of penicillin allergy testing. Patients with very low-risk penicillin allergies can safely undergo direct oral challenge (DOC) with a therapeutic dose of amoxicillin without prior skin testing. This study sought to establish a protocol for DOC in a primary care setting and test its efficacy and safety. Methods: A standardized DOC protocol was developed at 2 primary care sites in Southeast Michigan. Forty-nine patients across the 2 sites were identified as having very low-risk penicillin allergies and underwent DOC. Follow up phone calls were completed 1 week and 6 months following DOC. Results: All 49 patients had a negative DOC and successfully had their penicillin allergy delabeled from their electronic health record (EHR). No patients reported severe adverse reactions following DOC. All 22 patients who were successfully contacted 6 months after completing DOC reported willingness to take penicillin if prescribed in the future and believed they were no longer allergic to penicillin. Conclusions: The results suggest that DOC may be effectively and safely implemented in a primary care setting to delabel penicillin allergies in patients with low-risk penicillin allergies. This study may serve as a model to increase access to DOC for adults in rural settings or low-income patient populations with limited access to allergy specialists.

31 January 2024 in Research Square
Verified
Preprint
Open access
PDF file
ATAdam Tisch
Adam Tisch
SMSiddharth Madapoosi
Siddharth Madapoosi
SBStephen Blough
Stephen Blough
JRJan Rosa
Jan Rosa
SESean Eddy
Sean Eddy
FAFadhl Alakwaa
Fadhl Alakwaa
Background Single-cell RNA sequencing (scRNA-seq) and single-nucleus RNA sequencing (snRNA-seq) provide valuable insights into the cellular states of kidney cells. However, the annotation of cell types often requires extensive domain expertise and time-consuming manual curation, limiting scalability and generalizability. To facilitate this process, we tested the performance of five supervised classification methods for automatic cell type annotation. Results We analyzed publicly available sc/snRNA-seq datasets from five expert-annotated studies, comprising 62,120 cells from 79 kidney biopsy samples. Datasets were integrated by harmonizing cell type annotations across studies. Five different supervised machine learning algorithms (support vector machines, random forests, multilayer perceptrons, k-nearest neighbors, and extreme gradient boosting) were applied to automatically annotate cell types using four training datasets and one testing dataset. Performance metrics, including accuracy (F1 score) and rejection rates, were evaluated. All five machine learning algorithms demonstrated high accuracies, with a median F1 score of 0.94 and a median rejection rate of 1.8%. The algorithms performed equally well across different datasets and successfully rejected cell types that were not present in the training data. However, F1 scores were lower when models trained primarily on scRNA-seq data were tested on snRNA-seq data. Conclusions Our findings demonstrate that machine learning algorithms can accurately annotate a wide range of adult kidney cell types in scRNA-seq/snRNA-seq data. This approach has the potential to standardize cell type annotation and facilitate further research on cellular mechanisms underlying kidney disease.

19 November 2026 in Kidney360
Verified
Journal Article
Open access
NPNegiin Pourafshar
Negiin Pourafshar
SMSiddharth S. Madapoosi
Siddharth S. Madapoosi
AKAshkan Karimi
Ashkan Karimi
JTJeffrey Testani
Jeffrey Testani
CWChristopher S Wilcox
Christopher S Wilcox
Patients with heart failure often experience recurrent episodes of volume overload that respond to loop diuretics. However, this leads over time to neurohormonal activation, worsened sodium avidity, and tubular remodeling that contributes to diuretic resistance. Temporary diuretic withdrawal during volume removal by ultrafiltration, direct sodium removal, or other mechanical or pharmacological means may restore diuretic response at lower doses when therapy is re-introduced. This may persist for months after cessation of therapy. Collectively, these data support temporary periods of loop diuretic withdrawal or down-titration during management of volume loss by alternative means as a promising approach to overcoming diuretic resistance with heart failure.

24 July 2025 in American Journal of Nephrology
Verified
Journal Article
Closed access
SMSiddharth S. Madapoosi
Siddharth S. Madapoosi
LKLindsey M. Kornowske
Lindsey M. Kornowske
KDKenn B. Daratha
Kenn B. Daratha
CRChristina L. Reynolds
Christina L. Reynolds
CJCami R. Jones
Cami R. Jones
LMLaura H. Mariani
Laura H. Mariani
Introduction: Obesity is associated with chronic kidney disease (CKD) incidence and progression. We examined whether bariatric surgery is associated with change in eGFR trajectory among patients with and without CKD. Methods: Patients who underwent bariatric surgery at two health systems were identified using ICD-9/ICD-10 and CPT codes. Linear mixed models were fit on estimated glomerular filtration rate (eGFR) trajectory pre- and post-surgery among patients with or without CKD. Models were adjusted for age, sex, race, ethnicity, body mass index, hypertension, diabetes, follow-up duration, and type of bariatric surgery. Post-surgery, eGFR trajectory among patients with CKD was also compared following 1:2 propensity score matching to (1) patients without CKD who underwent surgery and (2) patients with CKD who did not undergo surgery. Results: Patients with CKD (n = 139) at Michigan Medicine had a slower annual rate of eGFR decline post-surgery compared to patients without CKD (n = 278) (1.54 [−2.26, −0.81] vs. 3.15 [−3.41, −2.87] mL/min/1.73 m2; p < 0.001), despite adjusting for degree of weight loss. Among patients with CKD, surgery was associated with a slower annual rate of eGFR decline (−0.20 [−0.83, 0.43] post-surgery vs. −1.11 [−1.37, −0.85] mL/min/1.73 m2 for non-surgery patients; p < 0.001). In an external validation study of patients with CKD in the Providence health system, bariatric surgery predicted an average increase in annualized eGFR slope by 1.19 [0.12, 2.25] mL/min/1.73 m2 (p = 0.03). Conclusion: Bariatric surgery is associated with less eGFR decline and may have weight-independent effects on preserving kidney function among persons living with obesity and CKD.