Ivan Lopez
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Ivan Lopez has a total of 193 co-authors. Here are the top co-authors by number of shared publications:
- Jonathan H. Chen
Stanford University
5 shared publications - Akshay Swaminathan
Department of Biomedical Data Science, Stanford University School of Medicine, Stanford, CA, USA
4 shared publications - Olivier Gevaert
Department of Biomedical Data Science Stanford University 94305 Stanford CA USA
3 shared publications - Stephen P. Ma
Department of Medicine, Stanford University School of Medicine, Stanford, CA, USA
3 shared publications - Ujwal Srivastava
Department of Computer Science, Stanford University
3 shared publications - Alexander Liang Ren
Department of Neurosurgery, Stanford University School of Medicine, Stanford, California
2 shared publications
Latest Publications
- Open Access
- 20 November 2026
- Published by Springer Nature in Research Square
Phenome-wide associations of CT-derived adipose area and density in 22,710 patients: uncovering unexpected disease connections
Adipose tissue plays a central role in metabolic and inflammatory homeostasis, yet its quantity and composition are rarely characterized together in routine care. We performed a phenome-wide association study linking abdominal CT scans from 22,710 patients at two tertiary care centers to electronic health record phenotypes. Subcutaneous and visceral fat area indices (SFI, VFI) and densities (SFD, VFD) were derived with a validated deep learning pipeline and tested against 624 phenotypes (2,496 tests). We identified 922 Bonferroni-significant associations across 60% (373/624) of phenotypes, recapitulating established relationships with morbid obesity (SFI odds ratio [OR] = 1.60), type 2 diabetes (VFI OR = 1.57), cirrhosis (VFD OR = 3.56) and pancreatic cancer (SFI OR = 0.56). Across the phenome, area and density odds ratios were inversely correlated (r = − 0.80), as expected for markers inversely correlated within each depot. When modelled jointly, density remained associated with disease independently of area for 90% (351/392) of phenotypes with a joint adipose association, whereas area was informative beyond density for 35% (136/392) and without density for 2% (6/392). These findings provide a hypothesis-generating atlas in which adipose density emerges as a candidate biomarker of systemic physiology. All analysis code and results are openly available (https://atesfet.github.io/fatwas.io).
- 20 November 2026
- Published by Springer Nature in Research Square
Open AccessPhenome-wide associations of CT-derived adipose area and density in 22,710 patients: uncovering unexpected disease connections
Adipose tissue plays a central role in metabolic and inflammatory homeostasis, yet its quantity and composition are rarely characterized together in routine care. We performed a phenome-wide association study linking abdominal CT scans from 22,710 patients at two tertiary care centers to electronic health record phenotypes. Subcutaneous and visceral fat area indices (SFI, VFI) and densities (SFD, VFD) were derived with a validated deep learning pipeline and tested against 624 phenotypes (2,496 tests). We identified 922 Bonferroni-significant associations across 60% (373/624) of phenotypes, recapitulating established relationships with morbid obesity (SFI odds ratio [OR] = 1.60), type 2 diabetes (VFI OR = 1.57), cirrhosis (VFD OR = 3.56) and pancreatic cancer (SFI OR = 0.56). Across the phenome, area and density odds ratios were inversely correlated (r = − 0.80), as expected for markers inversely correlated within each depot. When modelled jointly, density remained associated with disease independently of area for 90% (351/392) of phenotypes with a joint adipose association, whereas area was informative beyond density for 35% (136/392) and without density for 2% (6/392). These findings provide a hypothesis-generating atlas in which adipose density emerges as a candidate biomarker of systemic physiology. All analysis code and results are openly available (https://atesfet.github.io/fatwas.io).
- Open Access
- 18 February 2026
- Published by BMJ in BMJ Health & Care Informatics
- Vol. 33  (1) , e102007
Artificial intelligence translation in healthcare: an urgent call for evidence-informed policy frameworks
The deployment of artificial intelligence (AI) translation tools in healthcare is accelerating rapidly, yet regulatory frameworks lag dangerously behind clinical practice. Recent data reveal that 57% of US physicians are already using or planning to adopt AI translation services within the next year. This creates a critical policy vacuum where clinicians deploy tools with variable performance across languages, risking patient safety and deepening health inequities. We examine the fractured regulatory landscape, document performance disparities between well-resourced and digitally under-represented languages, and argue for an urgent, evidence-informed policy framework centred on patient comprehension rather than linguistic accuracy. We delineate a risk-stratified validation approach comprising two distinct tracks: a ‘Streamlined Pathway’ for tool-language combinations with robust existing evidence (eg, Spanish) and a ‘Standard Pathway’ requiring independent, prospective validation for digitally under-represented languages (eg, Haitian Creole). To ensure accountability, we propose establishing oversight bodies within the U.S. Department of Health and Human Services (HHS) or the Food and Drug Administration (FDA) to mandate pre-deployment validation and post-market monitoring. Without such action, AI translation risks creating a two-tier system where the 25.7 million Americans with non-English language preferences receive dramatically different care quality based solely on the language they speak.
- 18 February 2026
- Published by BMJ in BMJ Health & Care Informatics
- Vol. 33  (1) , e102007
Open AccessArtificial intelligence translation in healthcare: an urgent call for evidence-informed policy frameworks
The deployment of artificial intelligence (AI) translation tools in healthcare is accelerating rapidly, yet regulatory frameworks lag dangerously behind clinical practice. Recent data reveal that 57% of US physicians are already using or planning to adopt AI translation services within the next year. This creates a critical policy vacuum where clinicians deploy tools with variable performance across languages, risking patient safety and deepening health inequities. We examine the fractured regulatory landscape, document performance disparities between well-resourced and digitally under-represented languages, and argue for an urgent, evidence-informed policy framework centred on patient comprehension rather than linguistic accuracy. We delineate a risk-stratified validation approach comprising two distinct tracks: a ‘Streamlined Pathway’ for tool-language combinations with robust existing evidence (eg, Spanish) and a ‘Standard Pathway’ requiring independent, prospective validation for digitally under-represented languages (eg, Haitian Creole). To ensure accountability, we propose establishing oversight bodies within the U.S. Department of Health and Human Services (HHS) or the Food and Drug Administration (FDA) to mandate pre-deployment validation and post-market monitoring. Without such action, AI translation risks creating a two-tier system where the 25.7 million Americans with non-English language preferences receive dramatically different care quality based solely on the language they speak.
- Top Cited Papers
- 20 January 2026
- Published by Springer Nature in Nature Medicine
- Vol. 32  (3) , 943-951
Holistic evaluation of large language models for medical tasks with MedHELM
While large language models (LLMs) achieve near-perfect scores on medical licensing exams, these evaluations inadequately reflect the complexity and diversity of real-world clinical practice. Here we introduce MedHELM, an extensible evaluation framework with three contributions. First, a clinician-validated taxonomy organizing medical AI applications into five categories that mirror real clinical tasks—clinical decision support (diagnostic decisions, treatment planning), clinical note generation (visit documentation, procedure reports), patient communication (education materials, care instructions), medical research (literature analysis, clinical data analysis) and administration (scheduling, workflow coordination). These encompass 22 subcategories and 121 specific tasks reflecting daily medical practice. Second, a comprehensive benchmark suite of 37 evaluations covering all subcategories. Third, systematic comparison of nine frontier LLMs—Claude 3.5 Sonnet, Claude 3.7 Sonnet, DeepSeek R1, Gemini 1.5 Pro, Gemini 2.0 Flash, GPT-4o, GPT-4o mini, Llama 3.3 and o3-mini—using an automated LLM-jury evaluation method. Our LLM-jury uses multiple AI evaluators to assess model outputs against expert-defined criteria. Advanced reasoning models (DeepSeek R1, o3-mini) demonstrated superior performance with win rates of 66%, although Claude 3.5 Sonnet achieved comparable results at 15% lower computational cost. These results not only highlight current model capabilities but also demonstrate how MedHELM could enable evidence-based selection of medical AI systems for healthcare applications.
- 20 January 2026
- Published by Springer Nature in Nature Medicine
- Vol. 32  (3) , 943-951
Top Cited PapersHolistic evaluation of large language models for medical tasks with MedHELM
While large language models (LLMs) achieve near-perfect scores on medical licensing exams, these evaluations inadequately reflect the complexity and diversity of real-world clinical practice. Here we introduce MedHELM, an extensible evaluation framework with three contributions. First, a clinician-validated taxonomy organizing medical AI applications into five categories that mirror real clinical tasks—clinical decision support (diagnostic decisions, treatment planning), clinical note generation (visit documentation, procedure reports), patient communication (education materials, care instructions), medical research (literature analysis, clinical data analysis) and administration (scheduling, workflow coordination). These encompass 22 subcategories and 121 specific tasks reflecting daily medical practice. Second, a comprehensive benchmark suite of 37 evaluations covering all subcategories. Third, systematic comparison of nine frontier LLMs—Claude 3.5 Sonnet, Claude 3.7 Sonnet, DeepSeek R1, Gemini 1.5 Pro, Gemini 2.0 Flash, GPT-4o, GPT-4o mini, Llama 3.3 and o3-mini—using an automated LLM-jury evaluation method. Our LLM-jury uses multiple AI evaluators to assess model outputs against expert-defined criteria. Advanced reasoning models (DeepSeek R1, o3-mini) demonstrated superior performance with win rates of 66%, although Claude 3.5 Sonnet achieved comparable results at 15% lower computational cost. These results not only highlight current model capabilities but also demonstrate how MedHELM could enable evidence-based selection of medical AI systems for healthcare applications.
- Open Access
- 30 September 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8  (1) , 1-8
Operationalizing machine-assisted translation in healthcare
Over 25 million U.S. patients with a non-English language preference face unsafe care because discharge instructions and other materials are rarely translated in time. Advances in translation assisted by large language models can close this gap, but implementation guidance is scarce. Using the Consolidated Framework for Implementation Research, we outline key considerations—innovation, individuals, inner setting, implementation process, and outer setting—to offer healthcare leaders and policymakers a practical roadmap for language model machine-assisted translation integration.
- 30 September 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8  (1) , 1-8
Open AccessOperationalizing machine-assisted translation in healthcare
Over 25 million U.S. patients with a non-English language preference face unsafe care because discharge instructions and other materials are rarely translated in time. Advances in translation assisted by large language models can close this gap, but implementation guidance is scarce. Using the Consolidated Framework for Implementation Research, we outline key considerations—innovation, individuals, inner setting, implementation process, and outer setting—to offer healthcare leaders and policymakers a practical roadmap for language model machine-assisted translation integration.
- Open Access
- 23 September 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27  (1) , e76048
Fine-Tuning Methods for Large Language Models in Clinical Medicine by Supervised Fine-Tuning and Direct Preference Optimization: Comparative Evaluation
Background: Large language model (LLM) fine-tuning is the process of adjusting out-of-the-box model weights using a dataset of interest. Fine-tuning can be a powerful technique to improve model performance in fields like medicine, where LLMs may have poor out-of-the-box performance. The 2 common fine-tuning techniques are supervised fine-tuning (SFT) and direct preference optimization (DPO); however, little guidance is available for when to apply either method within clinical medicine or health care operations. Objective: This study aims to investigate the benefits of fine-tuning with SFT and DPO across a range of core natural language tasks in medicine to better inform clinical informaticists when either technique should be deployed. Methods: We use Llama3 8B (Meta) and Mistral 7B v2 (Mistral AI) to compare the performance of SFT alone and DPO across 4 common natural language tasks in medicine. The tasks we evaluate include text classification, clinical reasoning, text summarization, and clinical triage. Results: Our results found clinical reasoning accuracy increased from 7% to 22% with base Llama3 and Mistral2, respectively, to 28% and 33% with SFT, and then 36% and 40% with DPO (P=.003 and P=.004, respectively). Summarization quality, graded on a 5-point Likert scale, was 4.11 with base Llama3 and 3.93 with base Mistral2. Performance increased to 4.21 and 3.98 with SFT and then 4.34 and 4.08 with DPO (P<.001). F1-scores for provider triage were 0.55 for Llama3 and 0.49 for Mistral2, which increased to 0.58 and 0.52 with SFT and 0.74 and 0.66 with DPO (P<.001). F1-scores for urgency triage were 0.81 for Llama3 and 0.88 for Mistral2, which decreased with SFT to 0.79 and 0.87, and then experienced mixed results with DPO, achieving 0.91 and 0.85, respectively (P<.001 and P>.99, respectively). Finally, F1-scores for text classification were 0.63 for Llama3 and 0.73 for Mistral2, which increased to 0.98 and 0.97 with SFT, and then essentially did not change with DPO to 0.95 and 0.97, respectively (P=.55 and P>.99, respectively). DPO fine-tuning required approximately 2 to 3 times more compute resources than SFT alone. Conclusions: SFT alone is sufficient for simple tasks such as rule-based text classification, while DPO after SFT improves performance on the more complex tasks of triage, clinical reasoning, and summarization. We postulate that SFT alone is sufficient for simple tasks because SFT strengthens simple word-association reasoning, whereas DPO enables deeper comprehension because it is trained with both positive and negative examples, enabling the model to recognize more complex patterns. Ultimately, our results help inform clinical informaticists when to deploy either fine-tuning method and encourage commercial LLM providers to offer DPO fine-tuning for commonly used proprietary LLMs in medicine.
- 23 September 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27  (1) , e76048
Open AccessFine-Tuning Methods for Large Language Models in Clinical Medicine by Supervised Fine-Tuning and Direct Preference Optimization: Comparative Evaluation
Background: Large language model (LLM) fine-tuning is the process of adjusting out-of-the-box model weights using a dataset of interest. Fine-tuning can be a powerful technique to improve model performance in fields like medicine, where LLMs may have poor out-of-the-box performance. The 2 common fine-tuning techniques are supervised fine-tuning (SFT) and direct preference optimization (DPO); however, little guidance is available for when to apply either method within clinical medicine or health care operations. Objective: This study aims to investigate the benefits of fine-tuning with SFT and DPO across a range of core natural language tasks in medicine to better inform clinical informaticists when either technique should be deployed. Methods: We use Llama3 8B (Meta) and Mistral 7B v2 (Mistral AI) to compare the performance of SFT alone and DPO across 4 common natural language tasks in medicine. The tasks we evaluate include text classification, clinical reasoning, text summarization, and clinical triage. Results: Our results found clinical reasoning accuracy increased from 7% to 22% with base Llama3 and Mistral2, respectively, to 28% and 33% with SFT, and then 36% and 40% with DPO (P=.003 and P=.004, respectively). Summarization quality, graded on a 5-point Likert scale, was 4.11 with base Llama3 and 3.93 with base Mistral2. Performance increased to 4.21 and 3.98 with SFT and then 4.34 and 4.08 with DPO (P<.001). F1-scores for provider triage were 0.55 for Llama3 and 0.49 for Mistral2, which increased to 0.58 and 0.52 with SFT and 0.74 and 0.66 with DPO (P<.001). F1-scores for urgency triage were 0.81 for Llama3 and 0.88 for Mistral2, which decreased with SFT to 0.79 and 0.87, and then experienced mixed results with DPO, achieving 0.91 and 0.85, respectively (P<.001 and P>.99, respectively). Finally, F1-scores for text classification were 0.63 for Llama3 and 0.73 for Mistral2, which increased to 0.98 and 0.97 with SFT, and then essentially did not change with DPO to 0.95 and 0.97, respectively (P=.55 and P>.99, respectively). DPO fine-tuning required approximately 2 to 3 times more compute resources than SFT alone. Conclusions: SFT alone is sufficient for simple tasks such as rule-based text classification, while DPO after SFT improves performance on the more complex tasks of triage, clinical reasoning, and summarization. We postulate that SFT alone is sufficient for simple tasks because SFT strengthens simple word-association reasoning, whereas DPO enables deeper comprehension because it is trained with both positive and negative examples, enabling the model to recognize more complex patterns. Ultimately, our results help inform clinical informaticists when to deploy either fine-tuning method and encourage commercial LLM providers to offer DPO fine-tuning for commonly used proprietary LLMs in medicine.
- Open Access
- 22 September 2025
- Published by Oxford University Press (OUP) in Journal of the American Medical Informatics Association
- Vol. 32  (12) , 1865-1876
Predicting treatment retention in medication for opioid use disorder: a machine learning approach using NLP and LLM-derived clinical features
Objective: Building upon our previous work on predicting treatment retention in medications for opioid use disorder, we aimed to improve 6-month retention prediction in buprenorphine-naloxone (BUP-NAL) therapy by incorporating features derived from large language models (LLMs) applied to unstructured clinical notes. Materials and Methods: We used de-identified electronic health record (EHR) data from Stanford Health Care (STARR) for model development and internal validation, and the NeuroBlu behavioral health database for external validation. Structured features were supplemented with 13 clinical and psychosocial features extracted from free-text notes using the CLinical Entity Augmented Retrieval pipeline, which combines named entity recognition with LLM-based classification to provide contextual interpretation. We trained classification (Logistic Regression, Random Forest, XGBoost) and survival models (CoxPH, Random Survival Forest, Survival XGBoost), evaluated using Receiver Operating Characteristic-Area Under the Curve (ROC-AUC) and C-index. Results: XGBoost achieved the highest classification performance (ROC-AUC = 0.65). Incorporating LLM-derived features improved model performance across all architectures, with the largest gains observed in simpler models such as Logistic Regression. In time-to-event analysis, Random Survival Forest and Survival XGBoost reached the highest C-index (≈0.65). SHapley Additive exPlanations analysis identified LLM-extracted features like Chronic Pain, Liver Disease, and Major Depression as key predictors. We also developed an interactive web tool for real-time clinical use. Discussion: Features extracted using NLP and LLM-assisted methods improved model accuracy and interpretability, revealing valuable psychosocial risks not captured in structured EHRs. Conclusion: Combining structured EHR data with LLM-extracted features moderately improves BUP-NAL retention prediction, enabling personalized risk stratification and advancing AI-driven care for substance use disorders.
- 22 September 2025
- Published by Oxford University Press (OUP) in Journal of the American Medical Informatics Association
- Vol. 32  (12) , 1865-1876
Open AccessPredicting treatment retention in medication for opioid use disorder: a machine learning approach using NLP and LLM-derived clinical features
Objective: Building upon our previous work on predicting treatment retention in medications for opioid use disorder, we aimed to improve 6-month retention prediction in buprenorphine-naloxone (BUP-NAL) therapy by incorporating features derived from large language models (LLMs) applied to unstructured clinical notes. Materials and Methods: We used de-identified electronic health record (EHR) data from Stanford Health Care (STARR) for model development and internal validation, and the NeuroBlu behavioral health database for external validation. Structured features were supplemented with 13 clinical and psychosocial features extracted from free-text notes using the CLinical Entity Augmented Retrieval pipeline, which combines named entity recognition with LLM-based classification to provide contextual interpretation. We trained classification (Logistic Regression, Random Forest, XGBoost) and survival models (CoxPH, Random Survival Forest, Survival XGBoost), evaluated using Receiver Operating Characteristic-Area Under the Curve (ROC-AUC) and C-index. Results: XGBoost achieved the highest classification performance (ROC-AUC = 0.65). Incorporating LLM-derived features improved model performance across all architectures, with the largest gains observed in simpler models such as Logistic Regression. In time-to-event analysis, Random Survival Forest and Survival XGBoost reached the highest C-index (≈0.65). SHapley Additive exPlanations analysis identified LLM-extracted features like Chronic Pain, Liver Disease, and Major Depression as key predictors. We also developed an interactive web tool for real-time clinical use. Discussion: Features extracted using NLP and LLM-assisted methods improved model accuracy and interpretability, revealing valuable psychosocial risks not captured in structured EHRs. Conclusion: Combining structured EHR data with LLM-extracted features moderately improves BUP-NAL retention prediction, enabling personalized risk stratification and advancing AI-driven care for substance use disorders.
- Top Cited PapersOpen Access
- 19 January 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8  (1) , 1-11
Clinical entity augmented retrieval for clinical information extraction
Large language models (LLMs) with retrieval-augmented generation (RAG) have improved information extraction over previous methods, yet their reliance on embeddings often leads to inefficient retrieval. We introduce CLinical Entity Augmented Retrieval (CLEAR), a RAG pipeline that retrieves information using entities. We compared CLEAR to embedding RAG and full-note approaches for extracting 18 variables using six LLMs across 20,000 clinical notes. Average F1 scores were 0.90, 0.86, and 0.79; inference times were 4.95, 17.41, and 20.08 s per note; average model queries were 1.68, 4.94, and 4.18 per note; and average input tokens were 1.1k, 3.8k, and 6.1k per note for CLEAR, embedding RAG, and full-note approaches, respectively. In conclusion, CLEAR utilizes clinical entities for information retrieval and achieves >70% reduction in token usage and inference time with improved performance compared to modern methods.
- 19 January 2025
- Published by Springer Nature in npj Digital Medicine
- Vol. 8  (1) , 1-11
Top Cited PapersOpen AccessClinical entity augmented retrieval for clinical information extraction
Large language models (LLMs) with retrieval-augmented generation (RAG) have improved information extraction over previous methods, yet their reliance on embeddings often leads to inefficient retrieval. We introduce CLinical Entity Augmented Retrieval (CLEAR), a RAG pipeline that retrieves information using entities. We compared CLEAR to embedding RAG and full-note approaches for extracting 18 variables using six LLMs across 20,000 clinical notes. Average F1 scores were 0.90, 0.86, and 0.79; inference times were 4.95, 17.41, and 20.08 s per note; average model queries were 1.68, 4.94, and 4.18 per note; and average input tokens were 1.1k, 3.8k, and 6.1k per note for CLEAR, embedding RAG, and full-note approaches, respectively. In conclusion, CLEAR utilizes clinical entities for information retrieval and achieves >70% reduction in token usage and inference time with improved performance compared to modern methods.
- Open Access
- 25 October 2024
- Published by International Society of Global Health in Journal of Global Health
- Vol. 14, 04158
TeleHelp Ukraine: A distributed international telemedicine response to the ongoing war
The ongoing Russian war in Ukraine resulted in the displacement of vulnerable refugees and a catastrophic disruption in health care services within and beyond the nation’s borders, threatening the population’s right to care [1,2]. Within Ukraine, over 1000 health care facilities have been attacked since 24 February 2022, representing 15% of the country’s total health […]
- 25 October 2024
- Published by International Society of Global Health in Journal of Global Health
- Vol. 14, 04158
Open AccessTeleHelp Ukraine: A distributed international telemedicine response to the ongoing war
The ongoing Russian war in Ukraine resulted in the displacement of vulnerable refugees and a catastrophic disruption in health care services within and beyond the nation’s borders, threatening the population’s right to care [1,2]. Within Ukraine, over 1000 health care facilities have been attacked since 24 February 2022, representing 15% of the country’s total health […]
- Open Access
- 18 June 2024
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 13  (12) , 3555
Impact of Supine versus Prone Positioning on Segmental Lumbar Lordosis in Patients Undergoing ALIF Followed by PSF: A Comparative Study
Background: Anterior lumbar interbody fusion (ALIF) and posterior spinal fusion (PSF) play pivotal roles in restoring lumbar lordosis in spinal surgery. There is an ongoing debate between combined single-position surgery and traditional prone-position PSF for optimizing segmental lumbar lordosis. Methods: This retrospective study analyzed 59 patients who underwent ALIF in the supine position followed by PSF in the prone position at a single institution. Cobb angles were measured preoperatively, post-ALIF, and post-PSF using X-ray imaging. One-way repeated measures ANOVA and post-hoc analyses with Bonferroni adjustment were employed to compare mean Cobb angles at different time points. Cohen’s d effect sizes were calculated to assess the magnitude of changes. Sample size calculations were performed to ensure statistical power. Results: The mean segmental Cobb angle significantly increased from preoperative (32.2 ± 13.8 degrees) to post-ALIF (42.2 ± 14.3 degrees, Cohen’s d: −0.71, p < 0.0001) and post-PSF (43.6 ± 14.6 degrees, Cohen’s d: −0.80, p < 0.0001). There was no significant difference between Cobb angles after ALIF and after PSF (Cohen’s d: −0.10, p = 0.14). The findings remained consistent when Cobb angles were analyzed separately for single-screw and double-screw ALIF constructs. Conclusions: Both supine ALIF and prone PSF significantly increased segmental lumbar lordosis compared to preoperative measurements. The negligible difference between post-ALIF and post-PSF lordosis suggests that supine ALIF followed by prone PSF can be an effective approach, providing flexibility in surgical positioning without compromising lordosis improvement.
- 18 June 2024
- Published by MDPI AG in Journal of Clinical Medicine
- Vol. 13  (12) , 3555
Open AccessImpact of Supine versus Prone Positioning on Segmental Lumbar Lordosis in Patients Undergoing ALIF Followed by PSF: A Comparative Study
Background: Anterior lumbar interbody fusion (ALIF) and posterior spinal fusion (PSF) play pivotal roles in restoring lumbar lordosis in spinal surgery. There is an ongoing debate between combined single-position surgery and traditional prone-position PSF for optimizing segmental lumbar lordosis. Methods: This retrospective study analyzed 59 patients who underwent ALIF in the supine position followed by PSF in the prone position at a single institution. Cobb angles were measured preoperatively, post-ALIF, and post-PSF using X-ray imaging. One-way repeated measures ANOVA and post-hoc analyses with Bonferroni adjustment were employed to compare mean Cobb angles at different time points. Cohen’s d effect sizes were calculated to assess the magnitude of changes. Sample size calculations were performed to ensure statistical power. Results: The mean segmental Cobb angle significantly increased from preoperative (32.2 ± 13.8 degrees) to post-ALIF (42.2 ± 14.3 degrees, Cohen’s d: −0.71, p < 0.0001) and post-PSF (43.6 ± 14.6 degrees, Cohen’s d: −0.80, p < 0.0001). There was no significant difference between Cobb angles after ALIF and after PSF (Cohen’s d: −0.10, p = 0.14). The findings remained consistent when Cobb angles were analyzed separately for single-screw and double-screw ALIF constructs. Conclusions: Both supine ALIF and prone PSF significantly increased segmental lumbar lordosis compared to preoperative measurements. The negligible difference between post-ALIF and post-PSF lordosis suggests that supine ALIF followed by prone PSF can be an effective approach, providing flexibility in surgical positioning without compromising lordosis improvement.
- Open Access
- 1 June 2024
- Published by American Society of Clinical Oncology (ASCO) in JCO Clinical Cancer Informatics
- Vol. 8  (8) , e2300091
Extraction of Unstructured Electronic Health Records to Evaluate Glioblastoma Treatment Patterns
PURPOSE: Data on lines of therapy (LOTs) for cancer treatment are important for clinical oncology research, but LOTs are not explicitly recorded in electronic health records (EHRs). We present an efficient approach for clinical data abstraction and a flexible algorithm to derive LOTs from EHR-based medication data on patients with glioblastoma multiforme (GBM). METHODS: Nonclinicians were trained to abstract the diagnosis of GBM from EHRs, and their accuracy was compared with abstraction performed by clinicians. The resulting data were used to build a cohort of patients with confirmed GBM diagnosis. An algorithm was developed to derive LOTs using structured medication data, accounting for the addition and discontinuation of therapies and drug class. Descriptive statistics were calculated and time-to-next-treatment (TTNT) analysis was performed using the Kaplan-Meier method. RESULTS: Treating clinicians as the gold standard, nonclinicians abstracted GBM diagnosis with a sensitivity of 0.98, specificity 1.00, positive predictive value 1.00, and negative predictive value 0.90, suggesting that nonclinician abstraction of GBM diagnosis was comparable with clinician abstraction. Of 693 patients with a confirmed diagnosis of GBM, 246 patients contained structured information about the types of medications received. Of them, 165 (67.1%) received a first-line therapy (1L) of temozolomide, and the median TTNT from the start of 1L was 179 days. CONCLUSION: We described a workflow for extracting diagnosis of GBM and LOT from EHR data that combines nonclinician abstraction with algorithmic processing, demonstrating comparable accuracy with clinician abstraction and highlighting the potential for scalable and efficient EHR-based oncology research.
- 1 June 2024
- Published by American Society of Clinical Oncology (ASCO) in JCO Clinical Cancer Informatics
- Vol. 8  (8) , e2300091
Open AccessExtraction of Unstructured Electronic Health Records to Evaluate Glioblastoma Treatment Patterns
PURPOSE: Data on lines of therapy (LOTs) for cancer treatment are important for clinical oncology research, but LOTs are not explicitly recorded in electronic health records (EHRs). We present an efficient approach for clinical data abstraction and a flexible algorithm to derive LOTs from EHR-based medication data on patients with glioblastoma multiforme (GBM). METHODS: Nonclinicians were trained to abstract the diagnosis of GBM from EHRs, and their accuracy was compared with abstraction performed by clinicians. The resulting data were used to build a cohort of patients with confirmed GBM diagnosis. An algorithm was developed to derive LOTs using structured medication data, accounting for the addition and discontinuation of therapies and drug class. Descriptive statistics were calculated and time-to-next-treatment (TTNT) analysis was performed using the Kaplan-Meier method. RESULTS: Treating clinicians as the gold standard, nonclinicians abstracted GBM diagnosis with a sensitivity of 0.98, specificity 1.00, positive predictive value 1.00, and negative predictive value 0.90, suggesting that nonclinician abstraction of GBM diagnosis was comparable with clinician abstraction. Of 693 patients with a confirmed diagnosis of GBM, 246 patients contained structured information about the types of medications received. Of them, 165 (67.1%) received a first-line therapy (1L) of temozolomide, and the median TTNT from the start of 1L was 179 days. CONCLUSION: We described a workflow for extracting diagnosis of GBM and LOT from EHR data that combines nonclinician abstraction with algorithmic processing, demonstrating comparable accuracy with clinician abstraction and highlighting the potential for scalable and efficient EHR-based oncology research.