Citations

Clear

1134 citation(s) found for "Su, Yan".

Outcomes of Artificial Intelligence-Enhanced Colonoscopy in a Tertiary Clinical Setting
Nayar, Kapil D , Yakout, Abdelrahman , Sanchez, Patrick S., Haq, Khwaja S., Yuting, Huang, Conklin, Kerri, Wallace, Michael, Kumbhari, Vivek , Huang, Yuting , Bakheet, Nader , Sanchez, Patrick, Yan, Yan , Huang, Yuting , Conklin, Kerri , Osagiede, Osayande , Yan, Yan , Wallace, Michael B

Mayo Clinic

American Journal of Gastroenterology | February 12, 2026 | Volume: 121 | Issue: 4 | Pages: 974-981

10.14309/ajg.0000000000003953 | PMID 41677150 Get It

INTRODUCTION: Colonoscopies reduce colorectal cancer incidence and mortality, but challenges remain in detecting all precancerous lesions. Operator fatigue, technique variability, and subtle lesions can lead to missed adenomas, highlighting the need for tools that standardize detection quality. Computer-aided detection (CADe) systems use artificial intelligence to enhance lesion detection and improve screening colonoscopy quality. This study evaluates the impact of CADe on key quality and efficiency metrics in a clinical setting. METHODS: A single-center retrospective study at an academic tertiary center analyzed 4,028 colonoscopies from October 2022 to December 2023. The CADe system was used in 4 of 8 endoscopy suites, creating a CADe and control group. Propensity matching accounted for age, sex, indication, and physician. Primary outcomes included adenoma detection rate (ADR) and polyp detection rate (PDR). Adenomas per colonoscopy (APC) and polyps per colonoscopy (PPC) were also measured. Per-polyp analyses were performed as a tertiary outcome. RESULTS: ADR was significantly higher in the CADe group (38.6%) vs control (34.2%) (rate ratio [RR] = 1.127, P < 0.05). PDR was significantly higher in the CADe group (67.2%) vs control (59.4%) (RR = 1.130, P < 0.05). APC was significantly higher in the CADe group (M = 0.406, SD 0.637) vs control (M = 0.359, SD 0.599) (RR = 1.131, P < 0.05). PPC was significantly higher in the CADe group (M = 1.297, SD 1.188) vs control (M = 1.046, SD 1.002) (RR = 1.240, P < 0.05). The increase in polyp resection with CADe was driven predominantly by ≤5 mm lesions, with comparatively smaller differences for larger polyps. DISCUSSION: Implementing CADe in clinical practice significantly improved ADR, PDR, APC, and PPC. However, the increase in ADR seems inflated by the detection of polyps ≤5 mm, whose clinical significance in reducing colorectal cancer remains uncertain.
Redefining Upfront Resectable Pancreatic Ductal Adenocarcinoma: Should Vein Abutment Matter?
Tan, Po Hong, Thiels, Cornelius A., Poruk, Katherine E. , Miller, R., Hor, Hoe Yan , Tu, Melody Pi Yin , Huyser, Michelle R., Stucky, Chee-Chee, Wasif, Nabil , Bold, Richard J. , Smoot, R., Starlinger, Patrick P., Warner, Susanne G., Kendrick, M., Stauffer, John A., Truty, M., Fong, Zhi Ven, _, _, Poruk, Katherine, Miller, Richard Max, Thiels, Cornelius A. , Stucky, Chee‐Chee , Starlinger, Patrick , Warner, Susanne G. , Kendrick, Michael L. , Stauffer, John , Truty, Mark J. , Fong, Zhi Ven

Mayo Clinic

Journal of the National Comprehensive Cancer Network | February 9, 2026 | Volume: 24 | Issue: 2 | Pages: 27-33

10.6004/jnccn.2025.7097 | PMID 41671431 Get It

Background: Upfront resectable pancreatic ductal adenocarcinoma (PDAC) is defined by the NCCN as disease with ≤180° portomesenteric vein (PMV) abutment. However, the clinical significance of PMV abutment in this cohort remains unexplored, and we sought to investigate its impact. Methods: This multi-institutional study included patients who underwent pancreatoduodenectomy (PD) for upfront resectable PDAC from 2002 to 2023. Patients with PDAC with ≤180° PMV abutment were compared with those without. Overall survival (OS) was compared and stratified based on whether patients underwent upfront surgery or neoadjuvant therapy (NAT). Results: Among 1,446 patients included, 521 (36.0%) had PMV abutment. Median OS for patients with PMV abutment was 28.8 months compared with 31.2 months for those without ( P =.17). In a stratified analysis of patients who underwent upfront surgery, patients with PMV abutment had worse OS (19.2 vs 27.6 months; P <.05) and lower R0 resection rates (63.1% vs 87.2%; P <.05) compared with those without. However, in patients who underwent NAT, this difference in R0 resection rate was not observed (91.4% vs 94.0%, respectively; P =.27). Furthermore, there was no significant difference in median OS with NAT (42.5 vs 51.6 months, respectively; P =.48). Conclusions: In this largest cohort to date of patients with upfront resectable PDAC, PMV abutment was associated with worse OS compared with no PMV abutment when an upfront resection strategy was used. The use of NAT in patients with PMV abutment mitigated this survival difference with increased R0 resection rates. These findings suggest that patients with upfront resectable PDAC with PMV abutment benefit from a NAT approach for optimal survival outcomes, and that the current criteria guidelines need to be redefined.
Left ventricular ejection fraction and benefit of tricuspid valve interventions: Insights from the international TRIGISTRY
Heitzinger, Gregor, Dreyfus, J, Topilsky, Y, Bonis, M De, Carnero-Alcazar, M, Rudolph, V, Crestanello, J, Lurz, P H I L I P, Bax, J J, Maisano, F, Enriquez-Sarano, M, Bartko, P E, Messika-Zeitoun, D, Heitzinger, Gregor , Dreyfus, Julien , Dannenberg, Varius , Topilsky, Yan , Taramasso, Maurizio , Bonis, Michele De , Carnero, Manuel , Crestanello, Juan A. , Bax, J J B, Maisano, Francesco , Enriquez‐Sarano, Maurice , Bartko, P E, Messika–Zeitoun, David

Mayo Clinic

European Heart Journal | February 5, 2026 | Volume: 46 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehaf784.2363 Get It

Abstract Background The impact of treatment for tricuspid regurgitation (TR) across different levels of left ventricular ejection fraction (LVEF) remains uncertain. Purpose This study aimed to compare the outcomes of surgical and transcatheter tricuspid valve interventions (TTVI) to conservative (medical) management across all LVEF categories. Methods Patients with severe isolated TR from the TRIGISTRY, a multicenter international registry, were categorized based on LVEF (preserved EF-pEF:≥50%, mildly reduced EF-mrEF:41–49%, and reduced EF-rEF:≤40%). We assessed the impact of treatment modality and procedural success, defined as residual TR ≤mild-to-moderate, on two-year survival within each LVEF category. Results Among 2,384 patients, 1,383 had pEF, 400 had mrEF, and 601 had rEF. Compared to conservative management, surgery (P&lt;0.0005) and TTVI (P&lt;0.0001) were associated with a survival benefit in patients with pEF. No significant survival advantage was observed in patients with mrEF (P=0.28 for both), nor in those with rEF (P=0.76 and P=0.22 respectively). Similar results were obtained when surgical and transcatheter interventions were grouped together (P&lt;0.0001, P=0.17 and P=0.29 in patients with pEF, mrEF and rEF respectively). Patients with significant residual TR after TTVI exhibited a trend toward worse survival compared to those managed conservatively across all LVEF categories (P=0.47, P=0.33, and P=0.008 respectively)(Figure 1). Conclusion Tricuspid valve intervention, whether surgical or transcatheter-based, was associated with improved survival in patients with preserved LVEF but not in those with mildly reduced or reduced LVEF. Residual TR remained a significant prognostic factor across the entire LVEF spectrum. These findings highlight the need for careful patient selection when considering tricuspid interventions in individuals with reduced LVEF.Figure 1
Benefits of tricuspid valve intervention on outcomes based on underlying etiology in patients with severe functional tricuspid regurgitation: insights from the international TRIGISTRY
Iliadis, C, Dreyfus, J, Topilsky, Y, Bonis, M De, Carnero-Alcazar, M, Rudolph, V, Crestanello, J., Lurz, P H I L I P, Bartko, P E, Bax, J J, Enriquez-Sarano, M, Pfister, R, Messika-Zeitoun, D, Iliadis, Christos , Dreyfus, Julien , Topilsky, Yan , Taramasso, Maurizio , Bonis, Michele De , Carnero, Manuel , Bartko, P E, Bax, J J B, Maisano, Francesco , Enriquez‐Sarano, Maurice , Messika–Zeitoun, David

Mayo Clinic

European Heart Journal | February 5, 2026 | Volume: 46 | Issue: Supplement_1 | Pages: No page available

10.1093/eurheartj/ehaf784.2376 Get It

Abstract Background While tricuspid regurgitation (TR) is predominantly functional, it can arise from various cardiac conditions. This study aimed to evaluate whether the benefits of tricuspid valve intervention, compared to conservative management, differ based on the underlying etiology. Methods In the TRIGISTRY international cohort, which includes consecutive patients with severe isolated functional TR across 33 centers in 10 countries, we compared mortality rates at 2 years between patients who underwent a tricuspid valve intervention (surgical or transcatheter) versus conservative management according to TR etiology. Patients were classified into three groups: (1) Isolated TR (ITR), defined as left ventricular ejection fraction (LVEF) &gt;60%, systolic pulmonary artery pressure (sPAP) &lt;50 mmHg, and no prior left-heart intervention (LHI); (2) LHI-TR, comprising patients with a history of LHI; and (3) Undetermined TR (UTR), which included those who did not meet the criteria for ITR or LHI, as well as those with missing sPAP data. Given previous findings indicating that intervention provided no survival benefit for patients with a high TRI-SCORE, the present analysis was restricted to individuals with low to intermediate TRI-SCORE (≤5). Survival outcomes between groups were compared using restricted mean survival time (RMST). Results A total of 1,552 patients (72±13 years; 56% female) with severe functional TR and a low (48%) or intermediate (52%) TRI-SCORE were included. Of these, 772 (50%) underwent an intervention, while 780 (50%) were conservatively managed. TR was classified as ITR in 201 patients (13%), LHI in 395 (25%), and UTR in 956 (62%). Overall, patients with ITR exhibited the lowest comorbidity burden, whereas those with UTR had the highest. After adjusting for comorbidities and TRI-SCORE, a survival benefit of a tricuspid valve intervention was observed in patients with ITR (survival estimate: 0.17, 95% CI [0.009, 0.32], P=0.039) and, to a lesser extent, in patients with UTR (survival estimate: 0.09, 95% CI [0.0, 0.184], P=0.048), but not in those with prior LHI (survival estimate: 0.06, 95% CI [-0.08, 0.2], P=0.4). Similar results were found when excluding patients with moderate or greater residual TR after intervention. Conclusion In TRIGISTRY, patients with ITR appeared to derive the greatest benefit from tricuspid valve intervention, followed to a lesser extent by those with UTR, while no benefit was observed in patients with prior LHI. However, the potential influence of unaccounted confounders cannot be excluded, and further studies are needed to better identify patients’ subsets that may derive the most benefit from a tricuspid valve intervention.
Integrative Multi-Omics Analysis Identifies a Functional Enhancer Driving Tumorigenesis in Head and Neck Squamous Cell Carcinoma
Liu, Xiaorong, Pu, Jane Jingya, Lu, Xiaoling , Wang, Evelyn Yilin, Jiang, Qianhui , Yu, Dan , Yan, Bin, Zhong, Qi, Yang, Ping, Su, Yu-Xiong, Yu, Dan , Xu, Pengfei , Lee, Victor , Song, Jinlin , Tran, Nhan L. , Guan, Xinyuan , Wang, Yi-Xiang

Mayo Clinic

No container title available | February 5, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.64898/2026.02.03.703438 Get It

Abstract Enhancer is a critical epigenetic feature in head and neck squamous cell carcinoma (HNSCC), yet the functional roles of individual enhancers remain poorly understood. Here, we conducted an integrative multi-omics analysis based on publicly available ATAC-seq, H3K27ac ChIP-seq, transcriptomic profiling, and genetic association datasets to systematically map HNSCC-associated enhancers and their target genes. Integrating ATAC-seq and H3K27ac ChIP-seq, we identified 20,362 enhancers and 18,040 enhancer-associated genes, highlighting widespread regulatory complexity. Functional characterization of the TERT -associated enhancer GH05J001312 (GRCh38/hg38: chr5:1312099-1317743) revealed strong transcriptional activity in HNSCC. CRISPR-mediated deletion of a core sequence significantly reduced TERT expression, impaired cellular proliferation in vitro , and suppressed tumor growth in vivo , confirming its role as a key cis-regulatory element. RNA-seq analysis of enhancer-edited cells uncovered 742 differentially expressed genes enriched in cancer-related pathways, including MAPK and IL-17 signaling, indicating a broad transcriptional impact. Collectively, our findings establish GH05J001312 as a functional enhancer driving oncogenic programs in HNSCC and suggest enhancer-targeted strategies as a potential therapeutic avenue.
A human-in-the-loop explanation framework for morphologically transparent AI predictions from whole-slide images
Lou, Peiliang , Zhu, Yitan , Kumari, Roopa, Yang, William , Wang, Y. Claire , Guo, Ruifemg, Goode, Ellen, Huang, Yajue, Han, Wenchao, Feng, Tianshu , Wang, Chen, Zhu, Yitan , Chia, Nicholas , Yang, William, Wang, Yan, Novotny, Brenna C. , Winham, Stacey J. , Goode, Ellen L. , Han, Wenchao, Wang, Chen

Mayo Clinic

No container title available | January 30, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.64898/2026.01.27.701796 Get It

Abstract Deep learning models enable the prediction of clinical endpoints from whole-slide images (WSIs), but many such models function as “black boxes”, lacking transparency about whether and which histomorphological patterns drive their predictions, hindering interpretability and clinical adoption. Here we propose a human-in-the-loop explanation framework, MorphoXAI, which provides both local and global interpretability for deep learning models by incorporating human-expert interpretations. At the global level, it reveals the histomorphological patterns on which the model consistently relies to distinguish between classes of WSIs, as well as the patterns associated with confusion between classes. At the local level, it indicates which of these patterns are used in the prediction of an individual WSI and which regions within the slide correspond to such patterns. We validated our method on a deep learning model trained for ovarian tumor histologic subtype prediction. The results show that our framework generates explanations that accurately reflect the histomorphology underlying the model’s predictions at both global and local levels. For interpretability and clinical utility in diagnostic contexts, human evaluation results showed that our explanations were easy to interpret, rich in diagnostic features, and directly helpful for diagnostic decision-making, thereby enhancing pathologist-AI collaboration. Our work highlights that unifying global and local explanations and grounding them in expert-interpreted morphology enhances the interpretability and verifiability of deep learning models, thereby facilitating the transparent deployment of such models in clinical practice.
Prognostic factors for 90-day survival after stereotactic radiosurgery for brain metastasis patients
Loron, Ali Gharibi , Chow, William, Pumford, Andrew D., Zhao, Conan , Keough, Michael B. , You, Hyo Bin , Do, Bobby, Schwartz, Jenna , Foote, Robert L., Mahajan, Anita , Stien, Kathy J., Link, Michael J., Yan, Elizabeth Mary, Brown, Paul D., Pollock, Bruce E., Parney, I., Chow, William D, Zhao, Conan Y, Pumford, Andrew D. , Foote, Robert L. , Routman, David M. , Link, Michael J. , Burns, Terry C. , Yan, Elizabeth, Pollock, Bruce E.

Mayo Clinic

Journal of Neurosurgery | January 22, 2026 | Volume: 144 | Issue: 4 | Pages: 979-987

10.3171/2025.8.jns25859 | PMID 41569738 Get It

OBJECTIVE Brain metastases significantly impact neurocognitive function and overall survival. Stereotactic radiosurgery (SRS) is a cornerstone of treatment for patients with limited metastases and expected survival beyond 3 months. Despite current guidelines, up to 20% of patients with brain metastases undergoing SRS have been reported to die within 90 days. This study retrospectively evaluated prognostic factors associated with 90-day survival after SRS, aiming to improve patient selection. METHODS The authors retrospectively analyzed a cohort of 1546 patients who underwent Gamma Knife SRS for brain metastases at their institution between 2015 and 2023. One hundred seventy patients who survived less than 90 days after SRS were identified and case matched to 170 patients who survived over 90 days. Measured variables included patient demographic characteristics, tumor characteristics, treatment history, functional status, and control of the primary cancer. The authors modeled post-SRS 90-day survival using binomial and multivariate logistic regression. RESULTS Multivariate analysis highlighted Karnofsky Performance Score (KPS) < 70 (OR 17.4, p < 0.001), prior whole-brain radiation (OR 6, p = 0.004), and focal neurological deficits (OR 3.02, p = 0.003) as significant predictors of poor survival, whereas CNS progression before SRS (OR 0.22, p < 0.001) and control of systemic cancer (OR 0.556, p = 0.002) were associated with survival < 90 days. The predictive model demonstrated acceptable performance with an area under the curve (AUC) of 0.85, accuracy of 80%, sensitivity of 87%, and specificity of 71%. CONCLUSIONS Key predictors of 90-day survival after SRS for brain metastases include functional status (KPS), control of systemic cancer, CNS progression status, and focal neurological deficits. These findings are complementary factors that can assist in making decisions and SRS patient selection.
Establishing Reference Values in Healthy Participants for the Cardiac Troponin T High-Sensitivity Gen 6 Assay: REF-TSIX Global Reference Study
Daniels, Lori B. , Mueller, Christian, Giannitsis, Evangelos, Meex, Steven J. R., Buehlmann, David, Dilba, Peter, Bendig, Garnet, Cole, Mette, Body, Richard, Christenson, Robert H, deFilippi, Christopher R, Inoue, Kenji, Jaffe, A S, McCarthy, Cian P. , McCord, J., Neumann, Johannes T, Omland, T., Papendick, Cynthia , Sandoval, Yader, Twerenbold, Raphael, Peacock, W Frank, Mills, Nicholas L, investigators, the TSIX, Jacobs, Michael B., Lucksinger, Gregg, Powell, Richard, Johnson, Daniel, Coleman, Hugh, Fein, M, Surber, Joseph, Vu, Jack B., Hassman, Michael, Hernandez, John, Kirstein, Judith, Tomaszewski, Christian, Runyon, Michael S. , Gong, Jonathan , Wein, D., Andrabi, Sara , Kopin, David , Larrain, Germain, Tahirkheli, Naheem, Clemency, Brian, Abdullah, Shuaib, Hira, Ravi S. , Klausner, H A, Mahler, Simon A. , Crystal, Chad, Headden, Gary, Chaille, Peter, Cannon, Chad Aaron, Limkakeng, Alexander, Tolia, Vaishal , Knapp, Barry J., Massoomi, Mohammad, Sandoval, Yader, Ko, Emily, Luzzi, Verónica, Christenson, Robert, Carson, Jennifer Powers , Cacioppo, F , Kolev, E., Petrov, Andrey, Hazarbasanov, Dobri , Simova, lana, Magimaiseelan, Luci Suganthi, Kosch, Christine, Klein, G, Mehltretter, Kai, Sauter, Joachim, Baumann, Sybille, Moeckel, Martin, Twerenbold, Raphael, Muller-Hennessen, Matthias, Biener, Moritz , Findeisen, P., Omland, Torbjøm, Wilczek-Kosowska, Urszula, Zimmer-Satora, Ewa, Krecipro-Nizińska, Ewa, Miró, O, Díaz, Víctor Jiménez , Mueller, C, Christ, Michael, Lynch, Stephen, BODY, R. A. F., Matsuki, Shunji, Eto, Takashi, Tanaka, Takanori, Hanada, Ryuzo, Nakagami, Takuo, Nakama, T, Kadota, K., Jing, Shan, Yu, Shuwen, Zhang, Hongxuan, Zhang, Jun, Huo, Yong, Zhang, Yan, Yan, Li, Jiang, Longyuan, Chen, Weiqiang, Huang, Jianying, Zhao, Yan, Guan, Ling, Yang, Yanhua, Zhang, Qin, Qiu, Zhaohui, Body, Richard , Neumann, Franz–Josef , Müller, Christian , Mueller, Christian , Body, Rick , Eggers, Kai M. , Jaffe, Allan S. , Neumann, Johannes T , Sandoval, Yader , Tan, Jack Wei Chieh , Than, Martin , Twerenbold, Raphael , Mills, Nicholas L , Daniels, Lori B , Giannitsis, Evangelos , Meex, Steven J.R. , Christenson, Robert H. , Cobbaert, Christa M. , deFilippi, Christopher R. , McCord, James , Omland, Torbjørn , Sandoval, Yader , Tan, Jack Wei Chieh , Than, Martin , Twerenbold, Raphael , Surber, Joseph, Tomaszewski, Christian, Clemency, Brian, Headden, Gary , Cannon, Chad, Knapp, Barry , Tran, Nam K , Niederdöckl, Jan Daniel , Kolev, Emil, Popov, Todor, Sauter, Joachim, Baumann, Sybille , Möeckel, Martin , Müller‐Hennessen, Matthias, Findeisen, Peter, Scharnhorst, Volkher , Omland, T, Krecipro-Nizińska, Ewa, Christ, Michael , Nakagami, Takuo, Huo, Yong

Mayo Clinic

Clinical Chemistry | January 18, 2026 | Volume: 72 | Issue: 4 | Pages: 488-502

10.1093/clinchem/hvag011 | PMID 41564003 Get It

Abstract Background Measurement of cardiac troponin (cTn) using high-sensitivity assays is recommended for the diagnosis of myocardial infarction. We determined sex-specific and uniform 99th percentile upper reference limits (URLs) using the new Elecsys® Troponin T high-sensitivity Gen 6 assay in a global, healthy reference range cohort. Methods Lithium-heparin plasma and serum samples were prospectively collected from apparently healthy individuals aged ≥20 years across 34 global sites in the United States, Europe, China, and Japan. cTnT was measured using the Troponin T high-sensitivity Gen 6 assay on the Cobas® e 801 analyzer. Exclusion criteria were defined according to the 2022 IFCC guidance. Uniform and sex-specific 99th percentile URLs and nonparametric 95% CIs were determined in plasma and serum separately and combined. Results The final study population comprised 4147 participants (52.5% female) with a median (25th–75th percentiles) age of 48.0 (33.0–59.0) years; 45.8%, 47.9%, 5.2%, and 1.1% were White, Asian, Black, and other/unknown, respectively. For sample matrices combined (n = 8294), 81.0% and 99.2% of cTnT values were above the limit of detection in females and males, respectively. Sex-specific 99th percentile URLs (95% CI) were 18 (16–23) ng/L for females and 32 (28–35) ng/L for males; the uniform 99th percentile URL was 27 (24–31) ng/L. URLs were comparable in plasma and serum samples. Conclusions This study determined sex-specific and uniform 99th percentile URLs for the Troponin T high-sensitivity Gen 6 assay that were comparable irrespective of the matrix used in a large, global, healthy reference population.
Validation of 13 102 International Classification of Diseases, Tenth Revision, Clinical Modification codes using a large language model-based system
Wang, Yichen , Song, Yilin , Siu, Rex K. , Nimma, Induja R., Savage, Thomas , Wang, Yiming, Li, Zhichen, Ramai, Daryl, Wang, J., Badurdeen, Dilhana, Tao, Cui, Kumbhari, Vivek , Huang, Yuting, Siu, Rex, Huang, Yuting , Wang, Yichen , Yan, Yan , Tao, Cui

Mayo Clinic

Journal of the American Medical Informatics Association | January 14, 2026 | Volume: 33 | Issue: 5 | Pages: 947-956

10.1093/jamia/ocag008 | PMID 41670956 Get It

Abstract Objectives To comprehensively evaluate the validity of International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes for both prevalent diagnoses and less common diseases, and to assess the performance of a large language model (LLM)-based system in validating these codes. Materials and Methods This retrospective study analyzed hospital admissions from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. We developed a validated LLM-based system using GPT-4o, refined through iterative prompt engineering, to assess ICD-10-CM code validity. We measured the positive predictive value (PPV) of ICD-10-CM codes, PPV of principal and secondary diagnoses, and the performance of an LLM-based system in code validation. Results Among 865 079 assigned codes, the PPV was 84.6% (95% CI, 84.5%-84.6%). Principal diagnoses had a PPV of 93.9% (95% CI, 93.7%-94.1%), while secondary diagnoses had a PPV of 83.8% (95% CI, 83.7%-83.9%). The LLM system demonstrated high performance in validating ICD codes, achieving 93.6% accuracy, 95.4% sensitivity, and 85.2% specificity. Among correctly assigned secondary diagnoses, the majority (67.9%) represented historical or baseline conditions, while 32.1% reflected active conditions that deviated from baseline status; 22.3% of these emerged after hospital admission. PPV decreases with later diagnosis positions, with the largest decline occurring between principal and secondary diagnoses. Discussion and Conclusion In this large-scale evaluation, ICD-10-CM codes exhibited generally high accuracy, though variability existed by position and condition type. A validated LLM system performed comparably to physician review and offers a scalable means to improve coding accuracy. These findings support the potential for integrating LLM-based auditing into routine workflows to strengthen the quality of administrative and research data.
Self‐Reported Speech Outcomes in Oral Cavity Cancer
Baratz, Hannah , Renkert, Elisabeth, Li, Yan, McDaniels, James, Yeatts, Jennifer W., Yeoh, Melvyn, Montealegre, David Hernandez, Windon, Melina J. , Yin, Linda X , Orbelo, Diana, Yin, Linda X. , Ishikawa, Keiko , Orbelo, Diana M.

Mayo Clinic

Head & Neck | January 14, 2026 | Volume: 48 | Issue: 6 | Pages: 1559-1566

10.1002/hed.70134 | PMID 41532281 Get It

ABSTRACT Background Individuals with oral cavity cancer (OCC) have negatively impacted functional status and quality of life. Little is known regarding patients' perspectives of their speech impairments. This study compared self‐reported speech function between patients surgically treated for OCC and healthy controls using the Speech Handicap Index (SHI) and identified factors associated with worse scores. Methods This is a multiinstitutional cross‐sectional study of OCC patients and healthy controls recruited between July 2024 and March 2025. Participants completed the SHI and self‐reported speech quality. Demographics, tumor, and treatment details were abstracted from the electronic health record for OCC. Group differences were assessed with t ‐tests, χ 2 , and Fisher's exact tests. Results Among 69 OCC and 90 controls, OCC patients had significantly higher mean SHI scores (50.2 vs. 34.7, p < 0.01) and were less likely to rate their speech as “good” or “excellent” (47.8% vs. 85.6%, p < 0.01). Both speech and psychosocial domain scores were significantly worse among OCC. Free flap reconstruction was associated with worse scores (29.6 vs. 15.5, p = 0.02) and worse self‐rated speech quality. Conclusions Patients with surgically treated oral cavity cancer have significantly impaired self‐perceived speech and psychosocial functioning compared with healthy individuals. Greater deficits were observed in patients who required free flap reconstruction. Despite the functional deficit, nearly half of OCC patients reported satisfaction with their speech quality. These findings highlight the importance of using validated measures for speech function and support the need for early, multidisciplinary rehabilitation including speech therapy in this population, particularly for those undergoing free flap reconstruction.