Citations

Photoplethysmography (PPG) is a non-invasive optical technique commonly used to measure heart rate and oxygen saturation, but its waveform contains additional physiological information that can be analyzed using artificial intelligence (AI). This narrative review summarizes the emerging applications of AI-based PPG in cardiovascular, respiratory, sleep, hemodynamic, pregnancy-related, and portal-hypertension assessment, with the aim of evaluating its potential beyond conventional monitoring and identifying barriers to clinical translation. The literature search was conducted using PubMed, Google Scholar, IEEE Xplore, ScienceDirect, and SpringerLink. Additional relevant studies were identified through screening the reference lists of included articles. Studies published between 2002 and 2026 were identified to capture the development of PPG from conventional monitoring to newer AI-based applications. Human studies were prioritized, while relevant computational, simulated, synthetic, ex vivo, and technical studies were also included. Studies unrelated to PPG, duplicates, and studies with limited relevance were excluded. A total of 96 references were included, covering AI approaches such as convolutional and deep neural networks, ensemble methods, transfer learning, U-Net, generative adversarial networks, and Transformer-based models. Overall, the reviewed evidence suggests that AI-based PPG may support blood pressure estimation, atrial fibrillation detection, sleep and respiratory monitoring, vascular aging assessment, pulmonary hypertension screening, preeclampsia assessment, volume-status and compensatory-reserve assessment, and exploratory assessment related to portal hypertension. However, clinical translation remains limited by motion artifacts, sensor and measurement-site variability, skin-pigmentation-related bias, physiological and environmental influences, heterogeneous methods, limited external validation, and inconsistent clinical and regulatory standards.

Abstract Topic Benign Disease: Other Background Lichen planus is a chronic immune-mediated skin disease that may involve the esophagus. Esophageal lichen planus (ELP) is considered rare, but likely underdiagnosed. Lack of published guidelines for ELP results in significant treatment variability. Given heterogeneous practice patterns and emerging treatments, further outcome data are needed. Methods A retrospective cohort of ELP patients was assembled from 3 academic centers. Patients ≥ 18 years old were identified by ICD10 code search (L43.8) in the electronic health record, from 2006 to 2025. Patients were included if they had initial endoscopy, histology, and follow-up endoscopy showing features of LP, and excluded if there was diagnostic uncertainty, unavailable procedural documentation, or other esophageal diseases. Endoscopic findings were assessed qualitatively by manual review of reports. Data abstracted included demographics, presenting symptoms, endoscopic/histologic findings, treatment, and clinical outcomes. Statistical analysis was performed in R. Results 193 patients met inclusion criteria. Mean age at index endoscopy was 65 years, 86% were female, 98% were non-Hispanic Caucasian, 39% were ever smokers, and mean BMI was 26. Dysphagia was the most common presenting symptom (90%). Strictures were common (73%), 58% underwent dilation during initial endoscopy. Endoscopic findings included presence of sloughing (20%), exudate (22%) and friability (23%). Histology showed lymphocytic infiltrate (47%), parakeratosis/dyskeratosis (42%) and Civatte bodies (17%). Topical steroids (71%) and PPIs (63%) were used most frequently for treatment. Immunosuppressives (tacrolimus, systemic steroids, mycophenolate, methotrexate, hydroxychloroquine, cyclosporine, azathioprine, JAK inhibitor) were used in 54% of patients. Endoscopic improvement was seen in 66%, with highest response rates in tacrolimus (100%), hydroxychloroquine (100%), and a JAK inhibitor (100%). Symptom improvement was seen in 63%, with the greatest benefit seen in patients treated with a JAK inhibitor (100%) and hydroxychloroquine (86%). Conclusion In this large, multisite cohort, ELP demonstrated significant morbidity, with high rates of dysphagia, strictures, and inflammatory changes. Dilation appeared to be safe, with only a single iatrogenic perforation identified. Esophageal squamous cell cancer risk was low but present, at 4%. Despite heterogeneous treatment practices, several therapies – tacrolimus, hydroxychloroquine, and JAK inhibitors – were associated with marked endoscopic and symptomatic improvement. Prospective studies are needed to clarify the optimal approach to pharmacological management of ELP.

Portal hypertension is a central driver of morbidity and mortality in advanced chronic liver disease, yet its contribution to death is frequently underestimated owing to underreporting on death certificates. Population-level data on portal hypertension-related mortality trends and disparities in the United States remain limited. We conducted a retrospective, population-based mortality study using the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research Multiple Cause of Death database, identifying deaths with International Statistical Classification of Diseases and Related Health Problems, 10th Revision code K76.6 listed in any cause-of-death field from 1999 to 2023. Crude and age-adjusted mortality rates per 100,000 population were calculated and stratified by sex, U.S. Census region, race/ethnicity, and urban–rural status (the latter restricted to 1999–2020 due to data availability). Temporal trends were assessed using Joinpoint regression, with annual percent change (APC) reported as the primary trend metric and 95% confidence intervals presented for all estimates. A total of 34,551 portal hypertension-related deaths were identified. Overall age-adjusted mortality rates declined sharply from 1999 to 2002 (APC − 12.85%), plateaued for more than a decade, and rose thereafter. Mortality increased significantly in women after 2009 (APC + 6.69%) and in men after 2013 (APC + 6.18%). Among racial and ethnic groups, Hispanic or Latino individuals consistently demonstrated the highest mortality burden throughout the study period, followed by White individuals, whereas Black or African American individuals exhibited comparatively lower but significantly rising rates from 2006 onward. White individuals demonstrated the steepest recent acceleration, with a significant increase after 2014 (APC + 8.51%). Regionally, the South experienced the steepest post-2013 increase, while the Northeast retained the lowest absolute rates despite a significant rise after 2014. Urban populations showed early improvement through 2014 followed by rising mortality, whereas rural populations experienced a later but steeper rise beginning around 2013, with widening urban–rural disparities over time. Portal hypertension-related mortality has risen significantly since the mid-2010s across all demographic and geographic subgroups, with disproportionate burdens among Hispanic individuals, men, rural populations, and residents of the Southern United States. These findings highlight the urgent need for targeted prevention strategies, equitable access to hepatology care, and aggressive management of metabolic and alcohol-related risk factors to reverse this trajectory.

Abstract Interstitial lung disease (ILD) is a chronic, irreversible condition that is challenging to diagnose due to subtle symptoms and limitations of conventional tools. Noninvasive lung sound analysis using phonopulmography offers a promising alternative. This study evaluated the feasibility of a deep learning model for distinguishing normal and pathological lung sounds in ILD. Lung sound recordings were collected from 12 subjects (6 healthy, 6 ILD) and segmented beat-to-beat using simultaneously acquired ECG and phonopulmography, yielding 2,280 segments. Mel-frequency cepstral coefficients (MFCCs; 13 per segment) were extracted, and a one-dimensional CNN-based You Only Listen Once (YOLO) model was trained for segment classification. Evaluation used Leave-One-Out Cross-Validation, with patient-level classification by aggregating segment predictions via majority vote. On the test set (190 segments per class), the model achieved an average segment-level accuracy of 0.738 ± 0.087 and mean AUC of 0.850 ± 0.062. Precision, recall, and F1-score were 0.775 ± 0.148, 0.755 ± 0.183, and 0.738 ± 0.089, respectively. These preliminary results suggest that beat-level MFCCs with a 1D CNN-based YOLO model enable accurate segment- and patient-level classification. This feasibility study highlights the potential of AI-assisted auscultation for noninvasive monitoring and clinical decision support in ILD management, pending further validation in larger cohorts.

e23274 Background: Treatment-related symptoms substantially impair quality of life (QoL) and drive unplanned acute care use in oncology. Although patient-reported digital symptom monitoring is widely studied, outcomes remain inconsistent. We examined how intervention design influences clinical impact. Methods: We conducted a scoping systematic review of prospective randomized trials and pragmatic real-world studies evaluating patient-reported digital symptom monitoring or management interventions in adult oncology populations. Eligible studies reported symptom burden, QoL and/or acute care utilization. Data were abstracted on cancer type, digital modality, triage approach, and direction of outcomes. Protocols and qualitative studies were reviewed for context but excluded from outcome synthesis. Results: Across studies involving solid tumors and hematologic malignancies, interventions used web-based portals, mobile applications or interactive voice response systems. A consistent pattern emerged across the evidence base: interventions that paired real-time symptom reporting with structured, rule-based triage and proactive clinician response demonstrated improvements in symptom burden and QoL across multiple settings. In contrast, passive monitoring or self-management alone showed limited or no benefit despite similar reporting intensity. Effects on acute care utilization were heterogeneous, with reductions most often seen in pragmatic, workflow-integrated implementations. Conclusions: The clinical value of patient-reported digital symptom monitoring lies not in symptom collection itself, but in how reported data are reviewed and acted upon by clinicians. Monitoring alone is insufficient. Implementation-ready models that embed actionable triage and clinician response workflows are essential to improve patient outcomes and reduce avoidable utilization. Intervention approaches and clinical outcomes of digital symptom monitoring in oncology. Intervention Approach Digital Modality Actionable Triage / Clinician Response Symptom Burden Quality of Life Acute Care Use Monitoring with proactive clinician response Portal / App / IVR Yes (rule-based, clinician-reviewed) Improved Improved Reduced / Mixed Monitoring with automated alerts only Portal / App Partial Mixed Mixed NR Passive monitoring or self-management App / Portal No No benefit No benefit NR Pragmatic, workflow-integrated models Mixed Yes + staffing workflows Improved Improved Reduced Heterogeneous or non-integrated implementations Mixed Variable Mixed Mixed Mixed Abbreviations: IVR = Interactive Voice Response; QoL = Quality of Life; NR = Not Reported.

e13683 Background: Immune checkpoint inhibitors (ICIs) have transformed cancer treatment but are frequently complicated by immune-related adverse events (irAEs), which may result in treatment interruption, hospitalization and mortality. Reliable pre-treatment tools to identify patients at higher risk for irAEs, particularly checkpoint inhibitor pneumonitis (CIP) remain limited. Artificial intelligence (AI)-based prediction models have emerged but their clinical performance and readiness for implementation are unclear. Methods: We systematically searched PubMed, Embase, Scopus and Web of Science (inception-January 2025) for studies developing AI-based models to predict irAEs or CIP in patients receiving ICIs. Eligible studies included radiomics, deep learning, natural language processing and clinical machine-learning approaches. Two reviewers independently extracted data on cancer type, toxicity endpoint, modality, algorithm, validation strategy and model performance. Study design and reporting characteristics were assessed. Due to substantial methodological heterogeneity, results were synthesized descriptively. Results: Fifteen studies met inclusion criteria, including mixed solid tumor cohorts and lung-predominant populations. Predictive targets included any-grade irAEs and organ-specific toxicities, most commonly pneumonitis. Model approaches ranged from clinical and electronic health record–based models to CT- and PET/CT-based radiomics and multimodal imaging-clinical frameworks. Reported discriminatory performance varied widely, with AUC values ranging from 0.64 to > 0.90 across model types. The most consistent performance was observed in deep learning and multimodal radiomics-clinical models for pneumonitis, with AUCs frequently between 0.75 and 0.92, whereas several clinical-only models demonstrated more modest performance. Most studies were retrospective and single-center, used heterogeneous irAE definitions and lacked standardized imaging protocols. External validation and calibration assessment were uncommon. Conclusions: AI-based models for predicting irAEs in ICI-treated patients demonstrate promising performance, particularly for pneumonitis using radiomics and multimodal approaches. However, current evidence is limited by heterogeneity, scarce external validation and uncertain generalizability. Prospective, multi-institutional validation and standardized outcome definitions are needed before AI-driven irAE risk prediction can be integrated into routine oncology practice to guide surveillance and patient counseling.

Abstract Background: Hematologic toxicity, including early and prolonged neutropenia, anemia, and thrombocytopenia, is among the most common and clinically significant complications after CD19, CD20, CD22 and BCMA-directed CAR-T therapy. Several clinical scoresmost notably CAR-HEMATOTOX (CAR-HT) and the refined ALL-HThave been developed to predict post-infusion cytopenias, but their comparative performance across diseases has not been systematically evaluated. No prior review has summarized existing models or assessed the need for AI-based predictive approaches. Methods: We conducted a systematic review of studies that developed or validated predictive models or risk scores for hematologic toxicity after CAR-T therapy in hematologic malignancies. Eligible studies included clinical, cytokine-based, or composite predictors and reported discrimination metrics such as area under the curve (AUC). Extracted variables included CAR-T product, target antigen, disease type, predictors used, model type, validation strategy, and performance statistics. Results were synthesized qualitatively due to heterogeneity in model structure and outcome definitions. Results: Studies on the CAR-HT score in B-cell lymphomas and multiple myelomas reported AUC values of 0.89 and 0.82, respectively, for predicting severe or long-term neutropenia after CD19 and BCMA-directed CAR-T therapy. Studies on the ALL-HT score (a modified/refined version of CAR-HT replacing ferritin with bone-marrow burden to improve discrimination in B-ALL) reported AUC 0.84-0.90 for predicting prolonged neutropenia and worse overall survivability (OS) after CD19/CD22 CAR-T. A multicenter study including aggressive/indolent NHL, MM/PCL, and ALL patients treated with CD19, CD20, or BCMA CAR-T reported AUC 0.87 (eIPM-Pre) and 0.88 (eIPM-Post) for predicting ICA-HT. A CD19 CAR-T study in B-ALL and LBCL reported AUC 0.85 for an early hematotoxicity nomogram using TNF- and CRP. Conclusion: Existing prediction tools for hematologic toxicity after CAR-T therapy demonstrate strong but disease-specific performance. Yet inconsistent endpoints and heterogeneous predictors limit clinical generalizability. No current model incorporates contemporary AI or machine-learning methods, underscoring a significant unmet need. Advanced predictive frameworks may allow earlier identification of patients at risk for severe or prolonged cytopenias and guide more individualized supportive care, with the potential to reduce downstream complications. Citation Format: Mansha Gupta, Abhijith Vemulapalli, Swathi Cherukuri, Rithish Nimmagadda, Akhil Jain. Predictive models for hematologic toxicity after CAR-T therapy: A systematic review of current risk scores and unmet AI needs [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 5202.

Abstract Maternal obesity has been epidemiologically linked to an increased risk of sporadic colorectal cancer (CRC) in offspring, the third most prevalent cancer worldwide. This association underscores the lasting influence of early-life exposures on long-term health and disease susceptibility. Intestinal stem cells (ISCs), the long-lived cells that sustain epithelial renewal and often serve as the origin of intestinal cancers, are thought to play a central role in mediating this elevated cancer risk. However, how maternal obesity shapes the early-life programming and maturation of ISCs in their offspring remains poorly understood. Using mouse models of diet-induced obesity, we show that offspring exposed to a maternal high-fat Western diet (HFD) during pre- and postnatal development exhibit increased colonic proliferation, enhanced stem cell self-renewal, and a hypermetabolic state that persists into adulthood, even after dietary normalization. Alongside changes in stem cell properties, we observe a significant shift in epithelial cell composition, characterized by an increase in cells of secretory lineage at the expense of absorptive enterocytes, and these changes endure into adult life. Additionally, these changes are accompanied by an increased tumor burden in the loss of Apc heterozygosity model. Our data indicate that IL-17a, a pro-inflammatory cytokine, is critical in mediating these changes. Specifically, receptor IL-17RA/C expression is significantly upregulated in ISCs and secretory cells in maternal HFD-exposed offspring. Administration of IL-17a to intestinal organoids demonstrates sufficiency by mimicking the maternal obesogenic phenotype, indicating that immune-epithelial interactions influence ISC function and may be a driver of durable molecular patterning. These findings emphasize the long-term consequences of maternal HFD exposure on intestinal stem cell activity and epithelial composition, and possibly contribute to a higher risk of CRC in offspring later in life. Citation Format: Gourab Lahiri, Karla Mullen, Yesenia B. Millan, Swathi Sankar, Dominic R. Saiz, Thomas H. McDermott, Madeline Blatt, Abhigyan Shukla, Khashayarsha Khazaie, Miyeko D. Mana. Maternal obesity reprograms intestinal stem cells to elevate colorectal cancer risk in offspring [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 841.

ABSTRACT Maternal obesity is associated with increased risk of sporadic colorectal cancer (CRC) in offspring, suggesting that early-life environmental exposures durably shape disease susceptibility. Intestinal stem cells (ISCs), long-lived drivers of epithelial renewal and tumor initiation, are well poised to mediate this effect; however, how maternal obesity influences ISC programming during development remains poorly understood. Using mouse models of diet-induced obesity, we show that exposure to a maternal high-fat Western diet (mHFD) during pre- and postnatal development stably programs colonic ISCs. Offspring exhibit increased ISC proliferation, enhanced self-renewal, a hypermetabolic state, and altered epithelial lineage composition that persists into adulthood despite dietary normalization. These changes are accompanied by increased tumor burden following loss of Apc heterozygosity. Mechanistically, we identify the pro-inflammatory cytokine IL-17A as a key extrinsic driver and PPARd/a nuclear receptors as intrinsic mediators of the mHFD phenotype, revealing an immune-epithelial axis that programs ISC function during early life. Together, our findings demonstrate that maternal metabolic environments durably enhance stem cell fitness, providing a mechanistic link between developmental exposure and adult disease risk.

INTRODUCTION: The adjunctive use of p53 immunohistochemistry has been proposed as a potential tool to improve risk stratification in Barrett's esophagus (BE) with conflicting results. We performed a systematic review and meta-analysis to evaluate the performance of p53 in predicting progression to advanced neoplasia (high-grade dysplasia [HGD] and esophageal adenocarcinoma [EAC]) among patients with BE. METHODS: We searched multiple databases for studies that evaluated the use of aberrant p53 in esophageal biopsies among patients with BE and reported progression rates to HGD/EAC. The outcomes were p53 test characteristics and incidence and risk ratio (RR) for progression to HGD/EAC in patients with and without aberrant p53 using random effects models. RESULTS: Among the 27 included studies, the proportion of patients with aberrant p53 expression at baseline was 20% (95% CI: 14%, 27%). The rate of progression with and without aberrant p53 was 8 per 100 person-years (95% CI: 6, 11) and 0.3 per 100 person-years (95% CI: 0.1, 0.6), respectively. Compared with patients without aberrant p53 expression, those with aberrant p53 had a higher risk for progression to HGD/EAC in cohort studies (RR = 10.2 [95% CI: 6.9, 15.0]) and case control studies (RR = 3.3 [95% CI: 2.5, 4.4]). The sensitivity for progression for non-dysplastic Barrett's esophagus and low-grade dysplasia/indefinite for dysplasia was 42% and 77% and specificity was 96% and 76%, respectively. DISCUSSION: While aberrant p53 expression has been associated with an increased risk of progression to advanced neoplasia in patients with BE undergoing surveillance, the diagnostic characteristics are suboptimal precluding uniform clinical adoption. Prospective studies with standardized grading protocols are needed to determine whether p53-guided surveillance strategies can meaningfully improve patient outcomes.