About

Charitha Reddy is a Clinical Assistant Professor in the Department of Pediatrics, specializing in Cardiology at Stanford University. She is affiliated with the Center for Artificial Intelligence in Medicine & Imaging (AIMI) at Stanford. Her role involves integrating artificial intelligence and imaging technologies into medical research and clinical practice, contributing to the advancement of AI applications in healthcare. Her work focuses on leveraging AI to improve diagnostic and treatment strategies within pediatric cardiology, supporting innovative research and education in this interdisciplinary field.

Research topics

• Artificial Intelligence
• Computer Science
• Machine Learning
• Pediatrics
• Medicine
• Statistics
• Clinical psychology
• Computer network
• Distributed computing
• Internal medicine
• Data science
• Medical physics
• Radiology

Selected publications

• FINE: A Framework for Distributed Learning on Incomplete Observations for Heterogeneous Crowdsensing Networks

  International Journal of Advanced Research in Science Communication and Technology · 2023

  Senior authorCorresponding
  • Computer Science
  • Computer Science
  • Machine Learning

  Numerous crowdsensing applications have been developed recently in mobile social networks and vehicle networks. How to implement an accurate distributed learning process to estimate parameters of an unknown model in crowdsensing is a significant issue because centralised learning methods produce unreliable data gathering, expensive central servers, and privacy concerns. Due to this, we propose FINE, a distributed learning framework for imperfect data and non-smooth estimation, along with its design, analysis, and assessment. Our design, which is focused on creating a workable framework for learning parameters in crowdsensing networks accurately and efficiently, generalises earlier learning techniques by supporting heterogeneous dimensions of data records observed by various nodes as well as minimization based on non-smooth error functions.In particular, FINE makes use of a distributed dual average technique that efficiently minimises non-smooth error functions and a novel distributed record completion algorithm that enables each node to get the global consensus through effective communication with neighbours. All of these algorithms converge, as shown by our analysis, and the convergence rates are also obtained to support their efficacy. Through experiments on synthetic and actual networks, we assess how well our framework performs

  DOI

• 626: MODIFIED ASSESSMENT OF COMPETENCY IN THORACIC SONOGRAPHY (ACTS) SCALE IN THE NICU AND PICU

  Critical Care Medicine · 2022

  • Medicine
  • Medical physics
  • Radiology

  Introduction: Bedside point-of-care ultrasound (POCUS) is increasingly utilized in neonatal and pediatric populations. Standardized protocols to ensure high quality images for reliable interpretation are essential. However, there is a paucity of literature on assessing POCUS image quality. This project aims to adapt the adult Assessment of Competency in Thoracic Sonography (ACTS) Scale to pediatric and neonatal lung ultrasound (LUS) images. Methods: A multidisciplinary panel of experts from neonatology, pediatric intensive care, radiology, and cardiology identified the components essential for a good quality LUS in the pediatric and neonatal populations. The expert panel modified the adult ACTS scale for pediatric and neonatal populations. The modified scale scores the exam completeness, image labeling accuracy, image optimization, pathology detection, and agreement with the sonographer’s interpretation. Subsequently, the panel reviewers collected validity evidence for the scale by independently scoring 25 blinded LUS studies to determine concordance (Intraclass correlation (ICC), Cronbach’s Alpha, and percent agreement) between reviewers. Results: All LUS reviewed were performed for concerns of lung disease. We found a strong overall inter-rater reliability of the total ACTS score (ICC=0.689, Cronbach’s Alpha=0.930) across the reviewers. This was especially driven by a concordance in the image quality and completeness scoring (ICC=0.712, Cronbach’s Alpha = 0.937). There was high percent agreement of over 80% regarding the ability to interpret two of the four pathologies. Reviewers did not show strong agreement regarding sonographer’s interpretation (60.3%). Conclusions: Robust and reliable interpretation is based on high quality LUS images. This is the first study that defined essential components of a high-quality LUS image in pediatric and neonatal populations. Our modified pediatric and neonatal ACTS scale can reliably grade image quality. It may be an important quality assurance tool for a developing POCUS program. This study identified essential improvements for POCUS education. Further study is needed to delineate the image interpretation discrepancy.

  DOI

Frequent coauthors

• Bereketeab Haileselassie

  Stanford University

  2 shared

• Camille Hamilton

  University of California, Los Angeles

  1 shared

• Belinda Chan

  Chris O’Brien Lifehouse

  1 shared

• C Arpita

  Kurnool Medical College

  1 shared

• Sarah Hilgenberg

  Stanford University

  1 shared

• K J Chandan

  1 shared

• Reedhi Dasani

  1 shared

• Shazia Bhombal

  Children's Healthcare of Atlanta

  1 shared

Similar researchers at Stanford University

• Yi Cuih-287
• Zhenan Baoh-224
• Karl Deisserothh-218
• Michael Snyderh-215
• Roger Blandfordh-205