Charitha D. Reddy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Charitha D. Reddy has a total of 83 co-authors. Here are the top co-authors by number of shared publications:
- Michelle Kaplinski
Division of Cardiology, Department of Pediatrics, Stanford University School of Medicine, Lucile Packard Chidlren’s Hospital, Palo Alto, CA (M.K.).
4 shared publications - Rajesh Punn
Departments of Pediatrics, Lucile Packard Children’s Hospital Heart Center Clinical and Translational Research Program, Stanford University School of Medicine, CA. (R.Y., T.P., R.P., D.B.M.E.)
3 shared publications - Andrew Young Shin
Division of Cardiology, Lucile Packard Children's Hospital, Stanford University School of Medicine, Palo Alto, CA, United States
2 shared publications Bo Jin
HBI Solutions Inc., Palo Alto, CA, USA
2 shared publications- Brian Han
Division of Cardiology, Lucile Packard Children's Hospital, Stanford University School of Medicine, Palo Alto, CA, United States
2 shared publications Brice Gaudilliere
Stanford University School of Medicine, Stanford, CA, USA
2 shared publications
Latest Publications
- Open Access
- 10 February 2026
- Published by JMIR Publications Inc. in JMIR AI
- Vol. 5, e85221
Evaluating Large Language Model–Generated Clinical Summaries Through a Dual-Perspective Framework: Retrospective Observational Study
Large language models (LLMs) are increasingly used by patients and families to interpret complex medical documentation, yet most evaluations focus only on clinician-judged accuracy. In this study, 50 pediatric cardiac intensive care unit notes were summarized using GPT-4o mini and reviewed by both physicians and parents, who rated readability, clinical fidelity, and helpfulness. There were important discrepancies between parents and clinicians in the realm of helpfulness, along with important insights by clinicians assessing clinical accuracy and parents assessing readability. This study highlights the need for dual-perspective frameworks that balance clinical precision with patient understanding.
- 10 February 2026
- Published by JMIR Publications Inc. in JMIR AI
- Vol. 5, e85221
Open AccessEvaluating Large Language Model–Generated Clinical Summaries Through a Dual-Perspective Framework: Retrospective Observational Study
Large language models (LLMs) are increasingly used by patients and families to interpret complex medical documentation, yet most evaluations focus only on clinician-judged accuracy. In this study, 50 pediatric cardiac intensive care unit notes were summarized using GPT-4o mini and reviewed by both physicians and parents, who rated readability, clinical fidelity, and helpfulness. There were important discrepancies between parents and clinicians in the realm of helpfulness, along with important insights by clinicians assessing clinical accuracy and parents assessing readability. This study highlights the need for dual-perspective frameworks that balance clinical precision with patient understanding.
- Open Access
- 9 December 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8 (1) , 752
AI learning for pediatric right ventricular assessment: development and validation across multiple centers
Congenital and acquired heart disease affects approximately 1% of children worldwide, and right ventricular (RV) dysfunction is a common and complex manifestation in conditions such as congenital heart disease, pulmonary hypertension, and prematurity. Accurate RV assessment remains difficult due to the ventricle's irregular geometry and morphological variability in pediatric patients. Using 24,984 echocardiograms from 3993 children across four tertiary centers in North America and Asia, we developed and validated a video-based deep learning framework for automated RV functional assessment. The model performs frame-level ventricular segmentation and beat-by-beat estimation of fractional area change (FAC), classification of RV-related disease, and exploratory prediction of left ventricular ejection fraction (LV EF). A U²-Net architecture achieved high segmentation accuracy (Dice = 0.86 [A4C], 0.88 [PSAX]) and classification performance (AUC = 0.95 U.S., 0.97 Asia). In LV EF prediction, the model outperformed previous methods across cohorts. This validated framework enables expert-level, real-time quantification of pediatric ventricular function, enhancing diagnostic consistency, reducing manual workload, and supporting earlier intervention for children with heart disease, particularly in resource-limited settings.
- 9 December 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8 (1) , 752
Open AccessAI learning for pediatric right ventricular assessment: development and validation across multiple centers
Congenital and acquired heart disease affects approximately 1% of children worldwide, and right ventricular (RV) dysfunction is a common and complex manifestation in conditions such as congenital heart disease, pulmonary hypertension, and prematurity. Accurate RV assessment remains difficult due to the ventricle's irregular geometry and morphological variability in pediatric patients. Using 24,984 echocardiograms from 3993 children across four tertiary centers in North America and Asia, we developed and validated a video-based deep learning framework for automated RV functional assessment. The model performs frame-level ventricular segmentation and beat-by-beat estimation of fractional area change (FAC), classification of RV-related disease, and exploratory prediction of left ventricular ejection fraction (LV EF). A U²-Net architecture achieved high segmentation accuracy (Dice = 0.86 [A4C], 0.88 [PSAX]) and classification performance (AUC = 0.95 U.S., 0.97 Asia). In LV EF prediction, the model outperformed previous methods across cohorts. This validated framework enables expert-level, real-time quantification of pediatric ventricular function, enhancing diagnostic consistency, reducing manual workload, and supporting earlier intervention for children with heart disease, particularly in resource-limited settings.
- Open Access
- 7 October 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Evaluating Large Language Model–Generated Clinical Summaries Through a Dual-Perspective Framework: Retrospective Observational Study (Preprint)
UNSTRUCTURED Large language models (LLMs) are increasingly used by patients and families to interpret complex medical documentation, yet most evaluations focus only on clinician-judged accuracy. In this study, 50 pediatric cardiac intensive care unit notes were summarized using GPT-4o mini and reviewed by both physicians and parents, who rated readability, clinical fidelity, and helpfulness. There were important discrepancies between parents and clinicians in the realm of helpfulness, along with important insights by clinicians assessing clinical accuracy and parents assessing readability. This study highlights the need for dual-perspective frameworks that balance clinical precision with patient understanding.
- 7 October 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessEvaluating Large Language Model–Generated Clinical Summaries Through a Dual-Perspective Framework: Retrospective Observational Study (Preprint)
UNSTRUCTURED Large language models (LLMs) are increasingly used by patients and families to interpret complex medical documentation, yet most evaluations focus only on clinician-judged accuracy. In this study, 50 pediatric cardiac intensive care unit notes were summarized using GPT-4o mini and reviewed by both physicians and parents, who rated readability, clinical fidelity, and helpfulness. There were important discrepancies between parents and clinicians in the realm of helpfulness, along with important insights by clinicians assessing clinical accuracy and parents assessing readability. This study highlights the need for dual-perspective frameworks that balance clinical precision with patient understanding.
- Open Access
AI Learning for Pediatric Right Ventricular Assessment: Development and Validation Across Multiple Centers
Background: Congenital and acquired heart disease affects ∼1% of children globally, with right ventricular (RV) dysfunction being a common and complex issue due to conditions like congenital heart disease (CHD), pulmonary hypertension (PH), and prematurity. Accurate RV assessment is challenging due to its unique geometry, interventricular interactions, and morphological variability in pediatric patients. Fractional area change (FAC), a key echocardiographic measure, correlates strongly with disease severity, aiding in timely intervention and prognosis. AI learning shows the potential to automate and standardize RV assessments, overcoming traditional limitations and improving early diagnosis and management of pediatric cardiovascular disorders. Methods: Using 24,984 echocardiograms from 3,993 pediatric patients across four tertiary care centers (one in North America, three in Asia), we developed and validated an AI framework for automated RV assessment. The framework employs multi-task learning to perform ventricular segmentation, beat-by-beat quantification of RV FAC, and identification of cardiac abnormalities like PH. It was also extended to enhance left ventricular (LV) functional assessment. Findings: Our AI system achieved Dice similarity coefficients of 0.86 (apical-four-chamber, A4C) and 0.88 (parasternal-short-axis, PSAX) for RV segmentation, matching expert annotations. It demonstrated robust RV functional assessment, with AUCs of 0.95 (U.S. cohort) and 0.97 (Asian cohort). For PH classification, diagnostic accuracies were 0.95 (U.S.) and 0.94 (Asian), confirming consistent performance across populations. When extended to LV assessment, the framework significantly improved LV ejection fraction (EF) prediction in both U.S. and Asian cohorts. Interpretation: This validated AI framework enables reliable, automated ventricular function analysis, matching expert-level performance. By enhancing clinical workflows and standardizing pediatric cardiac assessments, it has the potential to improve care management for pediatric cardiovascular disorders, particularly in resource-limited settings. Funding: This work was supported by the U.S. NIH 1R41HL160362-01 to XBL and K23HL150279 to AT.
Open AccessAI Learning for Pediatric Right Ventricular Assessment: Development and Validation Across Multiple Centers
Background: Congenital and acquired heart disease affects ∼1% of children globally, with right ventricular (RV) dysfunction being a common and complex issue due to conditions like congenital heart disease (CHD), pulmonary hypertension (PH), and prematurity. Accurate RV assessment is challenging due to its unique geometry, interventricular interactions, and morphological variability in pediatric patients. Fractional area change (FAC), a key echocardiographic measure, correlates strongly with disease severity, aiding in timely intervention and prognosis. AI learning shows the potential to automate and standardize RV assessments, overcoming traditional limitations and improving early diagnosis and management of pediatric cardiovascular disorders. Methods: Using 24,984 echocardiograms from 3,993 pediatric patients across four tertiary care centers (one in North America, three in Asia), we developed and validated an AI framework for automated RV assessment. The framework employs multi-task learning to perform ventricular segmentation, beat-by-beat quantification of RV FAC, and identification of cardiac abnormalities like PH. It was also extended to enhance left ventricular (LV) functional assessment. Findings: Our AI system achieved Dice similarity coefficients of 0.86 (apical-four-chamber, A4C) and 0.88 (parasternal-short-axis, PSAX) for RV segmentation, matching expert annotations. It demonstrated robust RV functional assessment, with AUCs of 0.95 (U.S. cohort) and 0.97 (Asian cohort). For PH classification, diagnostic accuracies were 0.95 (U.S.) and 0.94 (Asian), confirming consistent performance across populations. When extended to LV assessment, the framework significantly improved LV ejection fraction (EF) prediction in both U.S. and Asian cohorts. Interpretation: This validated AI framework enables reliable, automated ventricular function analysis, matching expert-level performance. By enhancing clinical workflows and standardizing pediatric cardiac assessments, it has the potential to improve care management for pediatric cardiovascular disorders, particularly in resource-limited settings. Funding: This work was supported by the U.S. NIH 1R41HL160362-01 to XBL and K23HL150279 to AT.
- Open Access
Artificial Intelligence in Fetal and Pediatric Echocardiography
Echocardiography is the main modality in diagnosing acquired and congenital heart disease (CHD) in fetal and pediatric patients. However, operator variability, complex image interpretation, and lack of experienced sonographers and cardiologists in certain regions are the main limitations existing in fetal and pediatric echocardiography. Advances in artificial intelligence (AI), including machine learning (ML) and deep learning (DL), offer significant potential to overcome these challenges by automating image acquisition, image segmentation, CHD detection, and measurements. Despite these promising advancements, challenges such as small number of datasets, algorithm transparency, physician comfort with AI, and accessibility must be addressed to fully integrate AI into practice. This review highlights AI’s current applications, challenges, and future directions in fetal and pediatric echocardiography.
Open AccessArtificial Intelligence in Fetal and Pediatric Echocardiography
Echocardiography is the main modality in diagnosing acquired and congenital heart disease (CHD) in fetal and pediatric patients. However, operator variability, complex image interpretation, and lack of experienced sonographers and cardiologists in certain regions are the main limitations existing in fetal and pediatric echocardiography. Advances in artificial intelligence (AI), including machine learning (ML) and deep learning (DL), offer significant potential to overcome these challenges by automating image acquisition, image segmentation, CHD detection, and measurements. Despite these promising advancements, challenges such as small number of datasets, algorithm transparency, physician comfort with AI, and accessibility must be addressed to fully integrate AI into practice. This review highlights AI’s current applications, challenges, and future directions in fetal and pediatric echocardiography.
- 7 November 2023
- Published by Wolters Kluwer Health in Circulation
- Vol. 148 (Suppl_1)
Abstract 14080: Broadening Indications for Neonatal Balloon Atrial Septostomy in Dextro-Transposition of the Great Arteries Reduce Predictive Value of Fetal Echocardiography
Introduction: Neonates with dextro-transposition of the great arteries (d-TGA) require adequate mixing at the atrial level. If not present, these infants experience severe cyanosis, necessitating an urgent balloon atrial septostomy (BAS). Multiple studies have identified various fetal echocardiography (FE) findings as risk factors for postnatal BAS. Anecdotally, the predictive value of these findings remains poor in clinical use. Hypothesis: We aimed to determine the relationship between inter-rater reliability (IRR) of FE BAS predictors and need for postnatal BAS, hypothesizing that IRR would be poor and perhaps contributing to limited predictive value of FE findings. Methods: We identified d-TGA pregnancies followed at our institution between 2012 and 2022, excluding fetuses with large ventricular septal defects and pregnancies that were either terminated or lost to follow up. Three independent readers blinded to BAS outcome assessed 7 FE predictors: pulmonary vein Doppler velocity > 41 cm/sec, reverse flow in patent ductus arteriosus (PDA), restrictive PDA, restrictive patent foramen ovale (PFO), bidirectional PFO flow, flat septum primum, and hypermobile septum primum. IRR was determined with the S metric. Clinical information immediately before BAS was collected to evaluate postnatal physiology. A multivariable logistic regression was used to determine significant associations between FE findings and postnatal BAS. Results: Fifty-six fetuses were included in this study, of whom 25 (45%) underwent BAS. IRR was lowest for PFO flow direction and highest for restrictive PDA, likely due to its rarity. The remainder of FE predictors had either a moderate or substantial level of IRR. The last pre-BAS range of PaO 2 (24-49 mmHg), O 2 saturation (56-86%) and pH (7.18-7.52) reflected variable physiology. Of the 5 FE characteristics that could be included in the regression analysis, none were found to be significantly associated with odds of a BAS. Conclusions: Our results demonstrate substantial but not excellent IRR for current FE BAS predictors, but no significant association between these predictors and postnatal BAS in d-TGA patients. Poor predictive value of FE for postnatal BAS could be impacted by broadening indications for BAS.
- 7 November 2023
- Published by Wolters Kluwer Health in Circulation
- Vol. 148 (Suppl_1)
Abstract 14080: Broadening Indications for Neonatal Balloon Atrial Septostomy in Dextro-Transposition of the Great Arteries Reduce Predictive Value of Fetal Echocardiography
Introduction: Neonates with dextro-transposition of the great arteries (d-TGA) require adequate mixing at the atrial level. If not present, these infants experience severe cyanosis, necessitating an urgent balloon atrial septostomy (BAS). Multiple studies have identified various fetal echocardiography (FE) findings as risk factors for postnatal BAS. Anecdotally, the predictive value of these findings remains poor in clinical use. Hypothesis: We aimed to determine the relationship between inter-rater reliability (IRR) of FE BAS predictors and need for postnatal BAS, hypothesizing that IRR would be poor and perhaps contributing to limited predictive value of FE findings. Methods: We identified d-TGA pregnancies followed at our institution between 2012 and 2022, excluding fetuses with large ventricular septal defects and pregnancies that were either terminated or lost to follow up. Three independent readers blinded to BAS outcome assessed 7 FE predictors: pulmonary vein Doppler velocity > 41 cm/sec, reverse flow in patent ductus arteriosus (PDA), restrictive PDA, restrictive patent foramen ovale (PFO), bidirectional PFO flow, flat septum primum, and hypermobile septum primum. IRR was determined with the S metric. Clinical information immediately before BAS was collected to evaluate postnatal physiology. A multivariable logistic regression was used to determine significant associations between FE findings and postnatal BAS. Results: Fifty-six fetuses were included in this study, of whom 25 (45%) underwent BAS. IRR was lowest for PFO flow direction and highest for restrictive PDA, likely due to its rarity. The remainder of FE predictors had either a moderate or substantial level of IRR. The last pre-BAS range of PaO 2 (24-49 mmHg), O 2 saturation (56-86%) and pH (7.18-7.52) reflected variable physiology. Of the 5 FE characteristics that could be included in the regression analysis, none were found to be significantly associated with odds of a BAS. Conclusions: Our results demonstrate substantial but not excellent IRR for current FE BAS predictors, but no significant association between these predictors and postnatal BAS in d-TGA patients. Poor predictive value of FE for postnatal BAS could be impacted by broadening indications for BAS.
- Open Access
- 7 November 2023
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 12 (21) , e030083
Patterns of Aortic Dilation in Tetralogy of Fallot: An Analysis of 100 Fetal Echocardiograms Compared With Matched Controls
Background: Although aortic dilation is common in tetralogy of Fallot (TOF), its progression and risk of dissection are not well understood. The mechanism of dilation is primarily attributed to increased flow in utero; an alternative is unequal septation of the truncus arteriosus resulting in a larger aorta and inherently hypoplastic pulmonary artery (PA). If the latter is true, we hypothesize the aorta to PA ratio in TOF is stable throughout gestation, and sums of great artery dimensions are similar to controls. Methods and Results: We performed a single‐center retrospective study of fetuses with TOF (2014–2020) and matched controls. We compared sums of diameters, circumferences, and cross‐sectional areas of the aorta and PA and evaluated the aorta to PA ratio across gestation in 2 TOF subtypes: pulmonary stenosis and atresia (TOF‐PA). There were 100 echocardiograms with TOF (36% TOF‐PA) with median gestational age of 31 weeks (interquartile range 26.5–34.4) and median maternal age of 34 years (interquartile range 30–37). There were no differences in sums of great artery dimensions between TOF‐pulmonary stenosis and controls. In TOF‐PA, sums were significantly lower than controls ( P values r =0.08 [95% CI, −0.12 to 0.27], −0.06 [95% CI, −0.25 to 0.14]). Conclusions: The aorta in fetal TOF is large but grows proportionally throughout gestation, with sums of great artery dimensions similar to controls. TOF‐PA appears distinct from TOF‐pulmonary stenosis (with smaller sums), warranting further investigation. In conclusion, our findings suggest an intrinsic developmental mechanism contributes to aortic dilation in TOF.
- 7 November 2023
- Published by Wolters Kluwer Health in Journal of the American Heart Association
- Vol. 12 (21) , e030083
Open AccessPatterns of Aortic Dilation in Tetralogy of Fallot: An Analysis of 100 Fetal Echocardiograms Compared With Matched Controls
Background: Although aortic dilation is common in tetralogy of Fallot (TOF), its progression and risk of dissection are not well understood. The mechanism of dilation is primarily attributed to increased flow in utero; an alternative is unequal septation of the truncus arteriosus resulting in a larger aorta and inherently hypoplastic pulmonary artery (PA). If the latter is true, we hypothesize the aorta to PA ratio in TOF is stable throughout gestation, and sums of great artery dimensions are similar to controls. Methods and Results: We performed a single‐center retrospective study of fetuses with TOF (2014–2020) and matched controls. We compared sums of diameters, circumferences, and cross‐sectional areas of the aorta and PA and evaluated the aorta to PA ratio across gestation in 2 TOF subtypes: pulmonary stenosis and atresia (TOF‐PA). There were 100 echocardiograms with TOF (36% TOF‐PA) with median gestational age of 31 weeks (interquartile range 26.5–34.4) and median maternal age of 34 years (interquartile range 30–37). There were no differences in sums of great artery dimensions between TOF‐pulmonary stenosis and controls. In TOF‐PA, sums were significantly lower than controls ( P values r =0.08 [95% CI, −0.12 to 0.27], −0.06 [95% CI, −0.25 to 0.14]). Conclusions: The aorta in fetal TOF is large but grows proportionally throughout gestation, with sums of great artery dimensions similar to controls. TOF‐PA appears distinct from TOF‐pulmonary stenosis (with smaller sums), warranting further investigation. In conclusion, our findings suggest an intrinsic developmental mechanism contributes to aortic dilation in TOF.
- 15 September 2023
- Published by Elsevier
- p. 297-306
Artificial intelligence in pediatric congenital and acquired heart disease
- 15 September 2023
- Published by Elsevier
- p. 297-306
Artificial intelligence in pediatric congenital and acquired heart disease
- Open Access
- 13 June 2023
- Published by Cambridge University Press (CUP) in Cardiology in the Young
- Vol. 34 (1) , 171-177
Tricuspid valve annular tilt for assessment of pre- and post-intervention right ventricular volume in patients undergoing transcatheter pulmonary valve replacement
Introduction: Transcatheter pulmonary valve replacement serves as a successful alternative to surgical replacement of a right ventricle to pulmonary artery conduit. Guidelines for recommending transcatheter pulmonary valve replacement depend on MRI right ventricular volumes, which have been correlated to the echocardiographic measure of right ventricular annular tilt. We aim to assess whether right ventricular annular tilt can be a clinically useful alternative tool in the acute and long-term periods after transcatheter pulmonary valve replacement to assess right ventricular health. Methods: We reviewed 70 patients who underwent transcatheter pulmonary valve replacement at a single institution. Echocardiographic measurements were obtained prior to transcatheter pulmonary valve replacement, immediately after transcatheter pulmonary valve replacement, and within 6 months to 1 year after transcatheter pulmonary valve replacement. Right ventricular annular tilt measures the angle of the tricuspid valve plane relative to the mitral valve plane at end-diastole in the apical four-chamber view. Right ventricular fractional area change, right ventricular systolic strain, tissue Doppler velocity, and tricuspid annular plane systolic excursion Z-scores were obtained using published methods. Results: Right ventricular annular tilt decreased significantly immediately after transcatheter pulmonary valve replacement (p = 0.0004), and this reduction in right ventricular volume persisted at the mid-term follow-up (p < 0.0001). Fractional area change did not change significantly after transcatheter pulmonary valve replacement while right ventricular global strain improved at mid-term follow-up despite no significant difference immediately after transcatheter pulmonary valve replacement. Conclusions: Right ventricular annular tilt decreases both immediately after transcatheter pulmonary valve replacement and at mid-term follow-up. Right ventricular strain also improved after transcatheter pulmonary valve replacement, corresponding to the improved volume load. Right ventricular annular tilt can be considered as an additional echocardiographic factor to assess right ventricular volume and remodeling after transcatheter pulmonary valve replacement.
- 13 June 2023
- Published by Cambridge University Press (CUP) in Cardiology in the Young
- Vol. 34 (1) , 171-177
Open AccessTricuspid valve annular tilt for assessment of pre- and post-intervention right ventricular volume in patients undergoing transcatheter pulmonary valve replacement
Introduction: Transcatheter pulmonary valve replacement serves as a successful alternative to surgical replacement of a right ventricle to pulmonary artery conduit. Guidelines for recommending transcatheter pulmonary valve replacement depend on MRI right ventricular volumes, which have been correlated to the echocardiographic measure of right ventricular annular tilt. We aim to assess whether right ventricular annular tilt can be a clinically useful alternative tool in the acute and long-term periods after transcatheter pulmonary valve replacement to assess right ventricular health. Methods: We reviewed 70 patients who underwent transcatheter pulmonary valve replacement at a single institution. Echocardiographic measurements were obtained prior to transcatheter pulmonary valve replacement, immediately after transcatheter pulmonary valve replacement, and within 6 months to 1 year after transcatheter pulmonary valve replacement. Right ventricular annular tilt measures the angle of the tricuspid valve plane relative to the mitral valve plane at end-diastole in the apical four-chamber view. Right ventricular fractional area change, right ventricular systolic strain, tissue Doppler velocity, and tricuspid annular plane systolic excursion Z-scores were obtained using published methods. Results: Right ventricular annular tilt decreased significantly immediately after transcatheter pulmonary valve replacement (p = 0.0004), and this reduction in right ventricular volume persisted at the mid-term follow-up (p < 0.0001). Fractional area change did not change significantly after transcatheter pulmonary valve replacement while right ventricular global strain improved at mid-term follow-up despite no significant difference immediately after transcatheter pulmonary valve replacement. Conclusions: Right ventricular annular tilt decreases both immediately after transcatheter pulmonary valve replacement and at mid-term follow-up. Right ventricular strain also improved after transcatheter pulmonary valve replacement, corresponding to the improved volume load. Right ventricular annular tilt can be considered as an additional echocardiographic factor to assess right ventricular volume and remodeling after transcatheter pulmonary valve replacement.