Shawn Lyo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Shawn Lyo has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Misun Hwang
Department of Radiology, Children’s Hospital of Philadelphia, Philadelphia, PA 19104, USA
3 shared publications - Tessa Sundaram Cook
Department of Radiology, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA
3 shared publications - Alvand Hassankhani
Department of Radiology, Hospital of the University of Pennsylvania , Philadelphia, PA 19104
2 shared publications - Arastoo Vossough
Department of Radiology, Children’s Hospital of Philadelphia, Philadelphia, PA
2 shared publications - Colbey W. Freeman
Department of Radiology, University of Pennsylvania, Philadelphia, USA
2 shared publications - Luis Octavio Tierradentro-García
Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
2 shared publications
Latest Publications
- Open Access
- 3 September 2026
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- p. 1-13
Less Is More: Radiologists Prefer Brief Narratives in LLM-Generated Clinical History Summaries
Clinical histories are essential for accurate radiologic interpretation but are often lengthy and time-consuming to review amid increasing radiologist workloads. Large language models (LLMs) offer a potential solution by generating concise, clinically focused summaries from existing documentation; however, the optimal presentation format for radiologists remains unclear. This retrospective reader study evaluated the readability, efficiency, and radiologist preference for a two-stage LLM-based summarization pipeline designed to identify the optimal summary format for radiologic use. Ninety imaging studies across multiple modalities and care settings were processed to generate structured timeline summaries (stage 1) and brief narrative summaries (stage 2). Text length was reduced by approximately 86.8% and 95.2% for stage 1 and stage 2, respectively. Four radiologists independently rated representative subsets of cases for clarity, relevance, usefulness, and comprehensiveness and indicated their order of preference. Stage 2 summaries were preferred in 89% of ratings and were rated significantly higher than stage 1 for perceived clarity, relevance, and usefulness ( p < 0.0001), with equivalent comprehensiveness. Longer source documents and longer stage 1 summaries were associated with lower ratings for the detailed timeline format, whereas no such associations were observed for the brief narrative summaries. Manual review identified 25 total hallucinations across 90 cases, of which one was considered capable of altering image interpretation in a case with borderline radiographic features. These findings demonstrate that summary format, and particularly brevity, is critical for optimizing readability and radiologist preference. Automated summarization holds promise to reduce cognitive load and improve information accessibility in radiology practice.
- 3 September 2026
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- p. 1-13
Open AccessLess Is More: Radiologists Prefer Brief Narratives in LLM-Generated Clinical History Summaries
Clinical histories are essential for accurate radiologic interpretation but are often lengthy and time-consuming to review amid increasing radiologist workloads. Large language models (LLMs) offer a potential solution by generating concise, clinically focused summaries from existing documentation; however, the optimal presentation format for radiologists remains unclear. This retrospective reader study evaluated the readability, efficiency, and radiologist preference for a two-stage LLM-based summarization pipeline designed to identify the optimal summary format for radiologic use. Ninety imaging studies across multiple modalities and care settings were processed to generate structured timeline summaries (stage 1) and brief narrative summaries (stage 2). Text length was reduced by approximately 86.8% and 95.2% for stage 1 and stage 2, respectively. Four radiologists independently rated representative subsets of cases for clarity, relevance, usefulness, and comprehensiveness and indicated their order of preference. Stage 2 summaries were preferred in 89% of ratings and were rated significantly higher than stage 1 for perceived clarity, relevance, and usefulness ( p < 0.0001), with equivalent comprehensiveness. Longer source documents and longer stage 1 summaries were associated with lower ratings for the detailed timeline format, whereas no such associations were observed for the brief narrative summaries. Manual review identified 25 total hallucinations across 90 cases, of which one was considered capable of altering image interpretation in a case with borderline radiographic features. These findings demonstrate that summary format, and particularly brevity, is critical for optimizing readability and radiologist preference. Automated summarization holds promise to reduce cognitive load and improve information accessibility in radiology practice.
- 4 March 2026
- Published by Springer Nature in Pediatric Radiology
- Vol. 56 (6) , 1285-1294
RAD-AID International’s organized global impact on pediatric radiology
RAD-AID International is a non-profit organization improving radiology healthcare in medically underserved regions of the world through sustained partnerships and programmatic efforts targeting education, medical imaging technology advancement, and economic development. This paper examines the global landscape of pediatric radiology, emphasizing the significant disparities in low-resource areas. It details RAD-AID’s holistic strategy to customize programs according to local needs and enhance pediatric radiology through education, equipment, and technology. It also discusses the challenges and future directions for expanding pediatric radiology services, highlighting the importance of sustainability and the active involvement of local stakeholders.
- 4 March 2026
- Published by Springer Nature in Pediatric Radiology
- Vol. 56 (6) , 1285-1294
RAD-AID International’s organized global impact on pediatric radiology
RAD-AID International is a non-profit organization improving radiology healthcare in medically underserved regions of the world through sustained partnerships and programmatic efforts targeting education, medical imaging technology advancement, and economic development. This paper examines the global landscape of pediatric radiology, emphasizing the significant disparities in low-resource areas. It details RAD-AID’s holistic strategy to customize programs according to local needs and enhance pediatric radiology through education, equipment, and technology. It also discusses the challenges and future directions for expanding pediatric radiology services, highlighting the importance of sustainability and the active involvement of local stakeholders.
- 11 October 2025
- Published by Elsevier in Journal of the American College of Radiology
- Vol. 23 (3) , 342
Reply
- 11 October 2025
- Published by Elsevier in Journal of the American College of Radiology
- Vol. 23 (3) , 342
Reply
- Open Access
- 23 August 2025
- Published by Oxford University Press (OUP) in Radiology Advances
- Vol. 2 (5) , umaf029
Deep learning MRI halves scan time and preserves image quality across routine neuroradiologic examinations
Background: Magnetic resonance imaging (MRI) is a cornerstone of neuroimaging but is limited by lengthy acquisition times, which can lead to motion artifacts, patient discomfort, and delayed care. Deep learning reconstruction is an emerging technology that can offer image acquisition acceleration while maintaining image quality. Purpose: To compare image quality and acquisition efficiency between deep learning–accelerated vs conventional MRI (C-MRI) across a spectrum of routine neuroradiologic examinations. Materials and Methods: In this single-center retrospective study, 26 patients underwent imaging with a commercially available, FDA-cleared deep learning-accelerated MRI reconstruction algorithm (Deep Resolve, Siemens Healthineers), and C-MRI on a Siemens 3 T MAGNETOM Vida scanner between October 24 and November 14, 2023. A total of 113 sequence pairs were acquired across multiple body parts (brain [n = 28], cervical spine [n = 24], thoracic spine [n = 16], lumbar spine [n = 14], internal auditory canals [n = 5], sella [n = 5], neck [n = 5], temporomandibular joints [n = 6], brachial plexus [n = 4], and orbits [n = 6]) and sequences (T2 [n = 38], T1 [n = 30], short tau inversion recovery [n = 21], T1 post-contrast [n = 17], T2 fluid attenuated inversion recovery [n = 5], and proton density [n = 2]) and evaluated by 4 neuroradiologists blinded to the acquisition method for image quality using a 5-point Likert scale. Acquisition parameters were extracted from Digital Imaging and Communications in Medicine (DICOM) metadata and statistically compared. Rater preferences and interrater reliability were assessed using nonparametric tests and intraclass correlation coefficients. Results: Deep learning reduced mean scan time by 51.6% (95% CI: 45.7%–57.7%; from 110.8 seconds to 53.7 seconds; P < .001). Image quality assessments using a Likert scale showed scores slightly above neutral for signal-to-noise ratio (mean 3.51; 95% CI: 3.44–3.58), structural delineation (mean 3.51, 95% CI: 3.44–3.56), and overall image quality (mean 3.56, 95% CI: 3.49–3.63). However, poor interrater reliability (intraclass correlation [ICC] range: 0.06–0.33) showed that the observed differences were not consistent, indicating functional equivalence between conventional and deep learning images. Conclusion: Deep learning MRI enabled substantial scan time reductions while maintaining image quality.
- 23 August 2025
- Published by Oxford University Press (OUP) in Radiology Advances
- Vol. 2 (5) , umaf029
Open AccessDeep learning MRI halves scan time and preserves image quality across routine neuroradiologic examinations
Background: Magnetic resonance imaging (MRI) is a cornerstone of neuroimaging but is limited by lengthy acquisition times, which can lead to motion artifacts, patient discomfort, and delayed care. Deep learning reconstruction is an emerging technology that can offer image acquisition acceleration while maintaining image quality. Purpose: To compare image quality and acquisition efficiency between deep learning–accelerated vs conventional MRI (C-MRI) across a spectrum of routine neuroradiologic examinations. Materials and Methods: In this single-center retrospective study, 26 patients underwent imaging with a commercially available, FDA-cleared deep learning-accelerated MRI reconstruction algorithm (Deep Resolve, Siemens Healthineers), and C-MRI on a Siemens 3 T MAGNETOM Vida scanner between October 24 and November 14, 2023. A total of 113 sequence pairs were acquired across multiple body parts (brain [n = 28], cervical spine [n = 24], thoracic spine [n = 16], lumbar spine [n = 14], internal auditory canals [n = 5], sella [n = 5], neck [n = 5], temporomandibular joints [n = 6], brachial plexus [n = 4], and orbits [n = 6]) and sequences (T2 [n = 38], T1 [n = 30], short tau inversion recovery [n = 21], T1 post-contrast [n = 17], T2 fluid attenuated inversion recovery [n = 5], and proton density [n = 2]) and evaluated by 4 neuroradiologists blinded to the acquisition method for image quality using a 5-point Likert scale. Acquisition parameters were extracted from Digital Imaging and Communications in Medicine (DICOM) metadata and statistically compared. Rater preferences and interrater reliability were assessed using nonparametric tests and intraclass correlation coefficients. Results: Deep learning reduced mean scan time by 51.6% (95% CI: 45.7%–57.7%; from 110.8 seconds to 53.7 seconds; P < .001). Image quality assessments using a Likert scale showed scores slightly above neutral for signal-to-noise ratio (mean 3.51; 95% CI: 3.44–3.58), structural delineation (mean 3.51, 95% CI: 3.44–3.56), and overall image quality (mean 3.56, 95% CI: 3.49–3.63). However, poor interrater reliability (intraclass correlation [ICC] range: 0.06–0.33) showed that the observed differences were not consistent, indicating functional equivalence between conventional and deep learning images. Conclusion: Deep learning MRI enabled substantial scan time reductions while maintaining image quality.
- Open Access
- 15 July 2025
- Published by Elsevier in Journal of the American College of Radiology
- Vol. 22 (11) , 1271-1280
Integrating Large Language Models Into Radiology Education: An Interpretation-Centric Framework for Enhanced Learning While Supporting Workflow
- 15 July 2025
- Published by Elsevier in Journal of the American College of Radiology
- Vol. 22 (11) , 1271-1280
Open AccessIntegrating Large Language Models Into Radiology Education: An Interpretation-Centric Framework for Enhanced Learning While Supporting Workflow
- Open Access
- 1 July 2025
- Published by Elsevier in Journal of the American College of Radiology
- Vol. 22 (7) , 841-847
A Hitchhiker's Guide to Good Prompting Practices for Large Language Models in Radiology
- 1 July 2025
- Published by Elsevier in Journal of the American College of Radiology
- Vol. 22 (7) , 841-847
Open AccessA Hitchhiker's Guide to Good Prompting Practices for Large Language Models in Radiology
- 22 April 2025
- Published by American Society of Neuroradiology (ASNR) in American Journal of Neuroradiology
- Vol. 46 (10) , 1972-1984
An Atlas of Neonatal Neurovascular Imaging Anatomy as Depicted with Microvascular Imaging: The Intracranial Arteries
SUMMARY: Neurosonography is an invaluable diagnostic tool for assessing the neonatal brain and typically includes grayscale and conventional Doppler imaging. Microvascular imaging (MVI) is an emerging imaging technique that offers promising potential in evaluating neonatal intracranial pathology. MVI allows for sensitive and detailed assessment of neurovascular anatomy that can be challenging to visualize with other imaging modalities. This enhanced visualization not only improves the reliability in identifying neurovascular landmarks but also allows more accurate identification and assessment of adjacent parenchymal and extra-axial structures. Our review aims to offer an atlas of neurovascular structures as visualized with MVI, intended as a reference for future research in this field. Here, we focus specifically on the appearance of the intracranial arterial circulation as seen on MVI. Graphical Abstract
- 22 April 2025
- Published by American Society of Neuroradiology (ASNR) in American Journal of Neuroradiology
- Vol. 46 (10) , 1972-1984
An Atlas of Neonatal Neurovascular Imaging Anatomy as Depicted with Microvascular Imaging: The Intracranial Arteries
SUMMARY: Neurosonography is an invaluable diagnostic tool for assessing the neonatal brain and typically includes grayscale and conventional Doppler imaging. Microvascular imaging (MVI) is an emerging imaging technique that offers promising potential in evaluating neonatal intracranial pathology. MVI allows for sensitive and detailed assessment of neurovascular anatomy that can be challenging to visualize with other imaging modalities. This enhanced visualization not only improves the reliability in identifying neurovascular landmarks but also allows more accurate identification and assessment of adjacent parenchymal and extra-axial structures. Our review aims to offer an atlas of neurovascular structures as visualized with MVI, intended as a reference for future research in this field. Here, we focus specifically on the appearance of the intracranial arterial circulation as seen on MVI. Graphical Abstract
- 10 April 2025
- Published by American Society of Neuroradiology (ASNR) in American Journal of Neuroradiology
- Vol. 46 (10) , 1985-1993
An Atlas of Neonatal Neurovascular Imaging Anatomy as Depicted with Microvascular Imaging: The Intracranial Veins
Cerebral ultrasound is the preferred modality for evaluating the infant brain, offering real-time bedside assessment without exposure to ionizing radiation. Microvascular imaging ultrasound (MVI), an advanced Doppler technique, enables the visualization of microvascular slow-flow structures without the need for contrast agents. This innovative approach has the potential to provide valuable insights into the developing brain9s vasculature and hemodynamics. The utility of MVI has been investigated in various neurological conditions affecting neonates and infants, demonstrating its promise as a complementary tool to existing imaging modalities. By providing detailed, real-time images of the brain9s microvasculature, MVI can aid in the early detection, monitoring, and understanding of cerebral pathologies. In this review, we aim to present a comprehensive atlas of cerebral venous structures as visualized with MVI, serving as a reference for future research in this field and promoting the adoption of this technique in clinical practice for improved diagnosis and management of neurological conditions in the infant population. ABBREVIATIONS: MVI = Microvascular Imaging.
- 10 April 2025
- Published by American Society of Neuroradiology (ASNR) in American Journal of Neuroradiology
- Vol. 46 (10) , 1985-1993
An Atlas of Neonatal Neurovascular Imaging Anatomy as Depicted with Microvascular Imaging: The Intracranial Veins
Cerebral ultrasound is the preferred modality for evaluating the infant brain, offering real-time bedside assessment without exposure to ionizing radiation. Microvascular imaging ultrasound (MVI), an advanced Doppler technique, enables the visualization of microvascular slow-flow structures without the need for contrast agents. This innovative approach has the potential to provide valuable insights into the developing brain9s vasculature and hemodynamics. The utility of MVI has been investigated in various neurological conditions affecting neonates and infants, demonstrating its promise as a complementary tool to existing imaging modalities. By providing detailed, real-time images of the brain9s microvasculature, MVI can aid in the early detection, monitoring, and understanding of cerebral pathologies. In this review, we aim to present a comprehensive atlas of cerebral venous structures as visualized with MVI, serving as a reference for future research in this field and promoting the adoption of this technique in clinical practice for improved diagnosis and management of neurological conditions in the infant population. ABBREVIATIONS: MVI = Microvascular Imaging.
- Open Access
- 19 August 2024
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 38 (2) , 1265-1279
From Revisions to Insights: Converting Radiology Report Revisions into Actionable Educational Feedback Using Generative AI Models
Expert feedback on trainees’ preliminary reports is crucial for radiologic training, but real-time feedback can be challenging due to non-contemporaneous, remote reading and increasing imaging volumes. Trainee report revisions contain valuable educational feedback, but synthesizing data from raw revisions is challenging. Generative AI models can potentially analyze these revisions and provide structured, actionable feedback. This study used the OpenAI GPT-4 Turbo API to analyze paired synthesized and open-source analogs of preliminary and finalized reports, identify discrepancies, categorize their severity and type, and suggest review topics. Expert radiologists reviewed the output by grading discrepancies, evaluating the severity and category accuracy, and suggested review topic relevance. The reproducibility of discrepancy detection and maximal discrepancy severity was also examined. The model exhibited high sensitivity, detecting significantly more discrepancies than radiologists (W = 19.0, p < 0.001) with a strong positive correlation (r = 0.778, p < 0.001). Interrater reliability for severity and type were fair (Fleiss’ kappa = 0.346 and 0.340, respectively; weighted kappa = 0.622 for severity). The LLM achieved a weighted F1 score of 0.66 for severity and 0.64 for type. Generated teaching points were considered relevant in ~ 85% of cases, and relevance correlated with the maximal discrepancy severity (Spearman ρ = 0.76, p < 0.001). The reproducibility was moderate to good (ICC (2,1) = 0.690) for the number of discrepancies and substantial for maximal discrepancy severity (Fleiss’ kappa = 0.718; weighted kappa = 0.94). Generative AI models can effectively identify discrepancies in report revisions and generate relevant educational feedback, offering promise for enhancing radiology training.
- 19 August 2024
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 38 (2) , 1265-1279
Open AccessFrom Revisions to Insights: Converting Radiology Report Revisions into Actionable Educational Feedback Using Generative AI Models
Expert feedback on trainees’ preliminary reports is crucial for radiologic training, but real-time feedback can be challenging due to non-contemporaneous, remote reading and increasing imaging volumes. Trainee report revisions contain valuable educational feedback, but synthesizing data from raw revisions is challenging. Generative AI models can potentially analyze these revisions and provide structured, actionable feedback. This study used the OpenAI GPT-4 Turbo API to analyze paired synthesized and open-source analogs of preliminary and finalized reports, identify discrepancies, categorize their severity and type, and suggest review topics. Expert radiologists reviewed the output by grading discrepancies, evaluating the severity and category accuracy, and suggested review topic relevance. The reproducibility of discrepancy detection and maximal discrepancy severity was also examined. The model exhibited high sensitivity, detecting significantly more discrepancies than radiologists (W = 19.0, p < 0.001) with a strong positive correlation (r = 0.778, p < 0.001). Interrater reliability for severity and type were fair (Fleiss’ kappa = 0.346 and 0.340, respectively; weighted kappa = 0.622 for severity). The LLM achieved a weighted F1 score of 0.66 for severity and 0.64 for type. Generated teaching points were considered relevant in ~ 85% of cases, and relevance correlated with the maximal discrepancy severity (Spearman ρ = 0.76, p < 0.001). The reproducibility was moderate to good (ICC (2,1) = 0.690) for the number of discrepancies and substantial for maximal discrepancy severity (Fleiss’ kappa = 0.718; weighted kappa = 0.94). Generative AI models can effectively identify discrepancies in report revisions and generate relevant educational feedback, offering promise for enhancing radiology training.
- 1 July 2022
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 95 (1135) , 20211141
High-resolution neurosonographic examination of the lenticulostriate vessels in neonates with hypoxic-ischemic encephalopathy
Objective: To assess the feasibility of visualizing lenticulostriate vessels (LV) using a linear high-resolution ultrasound probe and characterize LV morphology to determine whether morphological alterations in LV are present in neonatal hypoxic–ischemic encephalopathy (HIE) as compared to the unaffected infants. Methods: We characterized LV by their echogenicity, width, length, tortuosity, and numbers of visualized stems/branches in neurosonographic examinations of 80 neonates. Our population included 45 unaffected (non-HIE) and 35 with clinical and/or imaging diagnosis of HIE. Of the neonates with clinical diagnosis of HIE, 16 had positive MRI findings for HIE (HIE+MRI) and 19 had negative MRI findings (HIE-MRI). Annotations were performed twice with shuffled data sets at a 1-month interval and intrarater reliability was assessed. Focused comparison was conducted between non-HIE, HIE+MRI and HIE-MRI neonates whose images were acquired with a high frequency linear transducer. Results: Studies acquired with the two most frequently utilized transducers significantly differed in number of branches (p = 0.002), vessel thickness (p = 0.007) and echogenicity (p = 0.009). Studies acquired with the two transducers also significantly differed in acquisition frequency (p < 0.001), thermal indices (p < 0.001) and use of harmonic imaging (p < 0.001). Groupwise comparison of vessels imaged with the most frequently utilized transducer found significantly fewer branches in HIE + MRI compared to HIE-MRI negative and non-HIE patients (p = 0.005). Conclusion: LV can be visualized in the absence of pathology using modern high-resolution neurosonography. Visualization of LV branches varies between HIE + MRI, HIE-MRI neonates and controls. Advances in knowledge: High-resolution neurosonography is a feasible technique to assess LV morphology in healthy neonates and neonates with HIE.
- 1 July 2022
- Published by Oxford University Press (OUP) in British Journal of Radiology
- Vol. 95 (1135) , 20211141
High-resolution neurosonographic examination of the lenticulostriate vessels in neonates with hypoxic-ischemic encephalopathy
Objective: To assess the feasibility of visualizing lenticulostriate vessels (LV) using a linear high-resolution ultrasound probe and characterize LV morphology to determine whether morphological alterations in LV are present in neonatal hypoxic–ischemic encephalopathy (HIE) as compared to the unaffected infants. Methods: We characterized LV by their echogenicity, width, length, tortuosity, and numbers of visualized stems/branches in neurosonographic examinations of 80 neonates. Our population included 45 unaffected (non-HIE) and 35 with clinical and/or imaging diagnosis of HIE. Of the neonates with clinical diagnosis of HIE, 16 had positive MRI findings for HIE (HIE+MRI) and 19 had negative MRI findings (HIE-MRI). Annotations were performed twice with shuffled data sets at a 1-month interval and intrarater reliability was assessed. Focused comparison was conducted between non-HIE, HIE+MRI and HIE-MRI neonates whose images were acquired with a high frequency linear transducer. Results: Studies acquired with the two most frequently utilized transducers significantly differed in number of branches (p = 0.002), vessel thickness (p = 0.007) and echogenicity (p = 0.009). Studies acquired with the two transducers also significantly differed in acquisition frequency (p < 0.001), thermal indices (p < 0.001) and use of harmonic imaging (p < 0.001). Groupwise comparison of vessels imaged with the most frequently utilized transducer found significantly fewer branches in HIE + MRI compared to HIE-MRI negative and non-HIE patients (p = 0.005). Conclusion: LV can be visualized in the absence of pathology using modern high-resolution neurosonography. Visualization of LV branches varies between HIE + MRI, HIE-MRI neonates and controls. Advances in knowledge: High-resolution neurosonography is a feasible technique to assess LV morphology in healthy neonates and neonates with HIE.