Citations

Clear

106 citation(s) found for "Helgeson, S A".

To Lengthen or Revise: Reevaluating Anterior Acetabular Overhang Thresholds in Iliopsoas Impingement After Total Hip Arthroplasty
Clark, David S , Yanamala, Sandeep R , Helgeson, Emma E , Berry, Daniel J , Levy, Bruce A , Taunton, Michael J

Mayo Clinic

Journal of Bone and Joint Surgery | October 9, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.2106/jbjs.26.00011 | PMID 42853909 Get It

Background: Iliopsoas impingement (IPI) is a known cause of persistent anterior hip pain following total hip arthroplasty (THA). Management of IPI has often relied on an 8-mm anterior acetabular overhang threshold to guide operative treatment, favoring acetabular revision for greater overhang and iliopsoas tenotomy for lesser overhang. These thresholds were initially established on the basis of the outcomes of open tenotomy, and now, with an improved understanding of IPI and the advent of arthroscopic iliopsoas fractional lengthening (IPFL), a review of the 8-mm threshold is warranted. Methods: A retrospective review was performed to identify patients who underwent arthroscopic IPFL following primary THA between 2017 and 2024, with a minimum follow-up of 1 year. Acetabular overhang was measured for all patients using cross-table lateral radiographs, and patients were stratified into <8-mm and ≥8-mm cohorts. Patient-reported outcome measures, reoperations, revisions, and complications were collected and analyzed on the basis of both the classic 8-mm threshold and continuous statistical methods. Results: We evaluated 86 hips undergoing IPFL. There were 83 patients (30 men and 53 women), with a mean age of 63.0 ± 11.2 years; with regard to race, 80 patients were White, 1 patient was Black or African American, and 2 patients listed other race or chose not to disclose. The acetabular overhang was <8 mm for 67 hips and ≥8 mm for 19 hips. The mean follow-up was 2.5 ± 1.6 years (range, 1 to 7.5 years). When comparing patients with <8-mm and ≥8-mm acetabular overhang, we observed no significant differences (p > 0.06) in the preoperative and postoperative visual analog scale (VAS) pain scores, modified Harris hip score, Single Assessment Numeric Evaluation (SANE) score, Tegner activity level, subjective groin pain improvement, subjective hip flexion strength improvement, and overall satisfaction. Reoperation (2.9% compared with 10.5%; p = 0.210) and acetabular revision rates (1.5% compared with 5%; p = 0.387) were not significantly different between the <8-mm and ≥8-mm overhang groups. Conclusions: Arthroscopic IPFL provided satisfactory outcomes for patients with IPI, independent of the magnitude of acetabular overhang. IPFL should be considered for first-line surgical management of IPI refractory to nonoperative management, even in the setting of substantial acetabular overhang, given positive outcomes and low rates of subsequent acetabular revision. Level of Evidence: Therapeutic Level IV . See Instructions for Authors for a complete description of levels of evidence.
Artificial Intelligence Enabled Dielectric Fingerprinting of Arrhythmogenic Substrates: A Precision Paradigm for Tissue Characterization Guided Cardiac Electrophysiology
Arunachalam, S P , Shariff, Mohammad Naveed , Kathirvel, Bargavi , Saraboji, Kaaviyashri , Kaur, Anmolpreet , Krishnappa, Manjunath, Elangovan, Poonguzhali , Jerold Wilson, Sancia Mary, Jayapradhaban Kala, Jayavinamika, Dey, Shuvashis , Mitra, Dipankar , Karuppiah, Shiva Sankari, Helgeson, Scott A , Tolkacheva, E G

Mayo Clinic

Sensors | September 25, 2026 | Volume: 26 | Issue: 19 | Pages: 6090

10.3390/s26196090 | PMID 42884291 Get It

Background: Electrical and dielectric properties of the myocardium depend on its composition, fibrosis, edema, ischemia, and heat damage, and hence may play an important role in defining the electrophysiologic substrate. Local impedance and other impedance-based methods give information about these parameters that is available to clinicians, but not direct or substitutable for intrinsic conductivity and permittivity of the tissue. The use of artificial intelligence has been thoroughly studied for the ECGs, electrograms, images, and computational data analysis, but not for dielectric or impedance measurements. Objective: The study aims to review the scientific basis of dielectric properties of heart muscle critically in cardiology and electrophysiology, to differentiate between existing clinical and conceptual applications of myocardial dielectrics, and to estimate current and potential contributions of artificial intelligence to their interpretation. Data Sources and Study Selection: The current work represents a structured narrative review in PubMed, Scopus, and Google Scholar databases from inception until May 2026 based on various combinations of keywords that refer to myocardial impedance, dielectric properties, conductivity, permittivity, local impedance, dielectric imaging, cardiac electrophysiology, catheter ablation, machine learning, and artificial intelligence. This review included experimental, ex vivo, animal, computational, and human studies related to myocardial dielectric properties and their use in electrophysiology. Evidence was classified according to four categories: direct human clinical evidence; preclinical/translation evidence; indirect evidence from related techniques; and extrapolation concept. As a result of heterogeneity between studies, sensing technologies, and outcomes, the results have been analyzed narratively. Key Findings: Evidence regarding the usage of AI algorithms, which have been primarily trained on myocardial dielectric or catheter-based local impedance data, is extremely scarce and mostly consists of preclinical or ex vivo data. On the contrary, the majority of clinically used AI algorithms in the field of electrophysiology make use of ECG data, intracardiac electrogram data, image processing, or computational modeling without any inputs related to dielectrics or local impedance. Conclusions and Implications for Practice: Dielectric sensing is an attractive supplement to traditional methods of voltage mapping and imaging, although the clinical readiness of dielectric sensing greatly varies depending on the application. At the present time, the existing data provide a basis for utilizing the local impedance measurement in the ablation procedure and evaluation of the myocardium properties, while the combination of artificial intelligence and dielectric sensing is still experimental. Before being introduced into the clinical electrophysiology practice, signal acquisition standardization and prospective multicenter trials are needed.
Pulmonary Function Test Order Redesign: A Quality Improvement Initiative
Sujanyal, Saurabh, Omercikoglu, Elcin , Alazab, Heba, Griffin, Angie, Koirala, Tapendra , Burger, Charles, Beagle, Lisa, Atkins, Shannon, Helgeson, Scott A. , Helgeson, Scott A , Alazab, Heba

Mayo Clinic

Journal of Patient Safety | September 16, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.1097/pts.0000000000001590 | PMID 42748326 Get It

Objectives: Pulmonary function tests (PFTs) are essential for diagnosis, management, and follow-up of a wide range of respiratory conditions as they provide quantitative insights into various aspects of lung function. Precision in ordering and scheduling PFTs, like any test, is crucial as errors can lead to misdiagnoses, delayed treatments, unnecessary retesting, inflated healthcare costs, and potentially compromised patient care. Hence, the aim of this quality improvement initiative was to reduce the PFT ordering error rate to below 5% by redesigning the ordering workflow and implementing a standardized order set tailored to specific clinical scenarios. Methods: The quality improvement initiative was implemented from August 2023 to March 2024. Data was collected from the electronic medical record (EMR) using predefined operational definitions to identify incorrect PFT orders and scheduling errors. A redesigned orderset was created to align with specific clinical presentations, including symptoms, suspected diagnosis, confirmed diagnosis, and presurgical evaluation. Scheduling was also restructured to match the expected testing time for each category. Education sessions were conducted with ordering providers, and ongoing feedback loops were established. Results: Among 575 preintervention PFT orders, 130 (22.6%) contained errors. Following implementation, review of 607 postintervention orders showed a reduction in error rates to 4.1%, representing an ∼80% improvement ( P <0.01). There was no increase in unnecessary testing, supporting the effectiveness of the intervention without unintended negative consequences. Conclusions: Implementation of a knowledge-based CDSS with standardized order sets significantly reduced PFT ordering errors and variability, improving guideline adherence and care quality. Continued refinement and integration with evolving technologies are essential for broader applicability.
Artificial Intelligence Enabled Diagnostics Using Photoplethysmography (PPG): Beyond SpO2 and Heart Rate
Joshi, Divya Dinesh, Saraboji, Kaaviyashri , Shaik, Asiya Tasleema , Sakalabaktula, Krishna Sai Kiran, Purohit, Shreya, Godugu, Swathi , Nirmal, Jasmine , Kayarkar, Riya, Correa, Bernardo Henrique Mendes , Gangidi, Namratha , Yadav, Pratibha, Yadav, Jyoti , Farook, Farshi , LEE, JIEUN , Shariff, Mohammed Naveed , Sundaram, Divaakar Siva Baala, Yerrapragada, Gayathri, Elangovan, Poonguzhali , Natarajan, T.K. , Janarthanan, Jayarajasekaran, Wilson, Sancia Mary Jerold, Karuppiah, Shiva Sankari, Abbasi, Zoma, Rapolu, Swetha, Bumrah, Simardeep Kaur, Bilgaiyan, Rashi, Premkumar, Chandra Rupini , Kaur, Anmolpreet , Pugalenthi, Lakshmi Sree , Sood, Divyanshi, Shivaram, Suganti, Iyer, Vivek , Vidhya, C. M., Masihi, Simin, Helgeson, Scott A. , Arunachalam, Shivaram P. , Natarajan, Thangeswaran , Jerold Wilson, Sancia Mary, Iyer, Vivek N , C M, Vidhya, Elangovan, Poonguzhali , Arunachalam, S P , Helgeson, Scott A , Shariff, Mohammad Naveed , Sakalabaktula, Krishna Sai Kiran , Godugu, Swathi , Yerrapragada, Gayathri, Bumrah, Simardeep Kaur, Bilgaiyan, Rashi , Shivaram, Suganti, Masihi, Simin

Mayo Clinic

Life | September 15, 2026 | Volume: 16 | Issue: 9 | Pages: 1530

10.3390/life16091530 | PMID 42795395 Get It

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.
Machine learning-based prediction of hypotension following emergency intubation
Sujanyal, Saurabh Ashok , Omercikoglu, Elcin , Alazab, Heba, Handral, Megha, LEE, JIEUN , Saunders, Hollie , Arunachalam, S P , Sujanyal, Saurabh , Alazab, Heba, Handral, Megha Sasidhar, Arunachalam, Shivaram P. , Helgeson, Scott A.

Mayo Clinic

BMJ Innovations | September 3, 2026 | Volume: No volume available | Issue: No issue available | Pages: bmjinnov-2026-001625

10.1136/bmjinnov-2026-001625 Get It

Introduction Post-intubation hypotension (PIH) is a frequent complication of emergency endotracheal intubation and is associated with increased mortality. Early identification of high-risk patients remains challenging, and existing prediction tools demonstrate moderate accuracy. Objective We aim to identify predictors of PIH and develop machine learning models for pre-intubation risk stratification. Methods We conducted a multicentre retrospective study across 3 Mayo Clinic hospitals, including 600 adults undergoing urgent intubation in the intensive care unit. PIH was defined by mean arterial pressure<65 mm Hg, systolic blood pressure (SBP)<80 mm Hg or ≥40% decline or vasopressor initiation/escalation within 30 min. Three model frameworks were developed: Shock Indices (SIs), Acute Physiology and Chronic Health Evaluation II (APACHE II) and Sequential Organ Failure Assessment (SOFA), incorporating demographic, haemodynamic, laboratory and sedation variables. Various machine learning algorithms were applied to analyse the three models. Results PIH occurred in 46% of patients included in the study. In the SIs model, stepwise logistic regression achieved an area under the curve (AUC) of 0.76. The APACHE II framework showed the strongest performance, with random forest achieving an ROC AUC of 0.82 and the ensemble model demonstrating the highest overall discrimination (Receiver operator Curve [ROC] AUC 0.83) with favourable accuracy and F1 metrics. The SOFA framework showed moderate performance, with an ensemble ROC AUC of 0.73. Key predictors included pre-intubation vasopressor use, SBP, platelet count and haematological parameters. Conclusions Machine learning models using routinely available pre-intubation data showed potential to predict PIH in this retrospective proof-of-concept study. These findings support further development of anticipatory artificial intelligence-driven risk stratification in critical care, but the models should not be considered ready for bedside implementation until they undergo prospective validation in larger, independent and more diverse cohorts.
The Impact of Hemipelvectomy and Hip Disarticulation on Lumbar Bone and Muscle Quality
Levy, Hannah A. , Messer, Caden, Ulrich, Marisa N. , Padilla, Sarah, Currier, Bradford, Nassr, Ahmad N., Freedman, Brett A., Helgeson, Melvin, Houdek, Matthew, Smither, Fantley, Sebastian, Arjun S. , Currier, Bradford L. , Nassr, Ahmad , Freedman, Brett A. , Helgeson, Melvin D. , Houdek, Matthew, Smither, F. Clay , Karamian, Brian A.

Mayo Clinic

Clinical Spine Surgery | August 28, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.1097/bsd.0000000000002133 | PMID 42663054 Get It

Study Design: Retrospective. Objective: To examine the long-term perioperative changes in lumbar bone quality, paraspinal muscle size and fat content, and facet degeneration after hemipelvectomy (HP) and hip disarticulation (HD). Summary of Background Data: Emerging evidence shows that transfemoral (TFA) and transtibial (TTA) amputations may accelerate lumbar modic changes and precipitate axial back pain. TFA and TTA amputation studies demonstrate that the severity of postamputation lower back pain is associated with the amputation type (TFA > TTA) and the degree of alteration in lumbopelvic kinematics. Methods: All patients who underwent HP or HD for malignancy or infection at an academic center between 2003 and 2022 were retrospectively identified. Preoperative to postoperative changes in multifidus, erector spinae, quadratus lumborum, and psoas cross-sectional area (CSA), and Hounsfield units (HU) were calculated on CT for the amputated and contralateral side at the level of the L3–4 disc space. Paired univariate analysis was utilized to compare preoperative and postoperative outcomes. Results: A total of 16 patients (8 HP, 8 HD) met the inclusion/exclusion criteria. The preoperative to postoperative changes in HUs in the L1, L3, and L5 vertebral bodies were −27.0 ( P =0.004), −51.5 ( P =0.002), and −48.1 ( P =0.001), respectively. The decrease in L1 HU was greater for prosthesis nonusers (user: −16.50, nonuser: −51.32, P =0.021). L5–S1 segment facet degeneration increased postoperatively ( P =0.030, n=7 progressed). There were significant preoperative to postoperative changes in psoas CSA (preop: 1063.4, postop: 586.5, P =0.005) and psoas HU (preop: 50.95, postop: 33.55, P <0.001) on the amputated but not the contralateral side. Conclusion: Hind-quarter amputations were associated with decreased lumbar bone quality, progression of facet degeneration, and loss of amputated side psoas muscle tone. Daily prosthesis use was moderately protective for lumbar bone quality following amputation. HP but not HD was associated with postoperative loss of amputated side quadratus lumborum tone. Level of Evidence: Level III.
Mortality risk after traumatic fractures in autofused spines from ankylosing spondylitis or DISH: age may matter more than treatment
Pennington, Zach, El-Gohary, Mohamed M., Lakomkin, Nikita, Ransom, Seth, Dennison, Stanley, Obiri-Yeboah, Derrick, Martini, Michael L., Kendall, Nicholas, Gharibi, Orod , Owens, Virgenal, Krauss, William E. , Helgeson, Melvin D. , Sebastian, Arjun S. , Mikula, Anthony L. , Fogelson, Jeremy L. , Freedman, Brett A., Elder, Benjamin D., Pennington, Zach , Pennington, Zach , Hamouda, Abdelrahman M. , Lakomkin, Nikita , Ransom, Seth C. , Owens, Virgenal, Clarke, Michelle J. , Nassr, Ahmad , Freedman, Brett Arthur , Elder, Benjamin D.

Mayo Clinic

Journal of Neurosurgery: Spine | August 26, 2026 | Volume: No volume available | Issue: No issue available | Pages: 1-11

10.3171/2026.4.spine251776 | PMID 42664512 Get It

OBJECTIVE Multisegmental pathological autofusion is a hallmark feature of ankylosing spondylitis (AS) and diffuse idiopathic skeletal hyperostosis (DISH), creating a rigid, brittle spine highly susceptible to three-column fractures and devastating neurological injury. This study aimed to evaluate survival outcomes in this patient population and identify clinical and demographic predictors of mortality. METHODS Adults with DISH or AS who sustained a fracture and were admitted to a single tertiary care center for spinal fracture were retrospectively identified using ICD-10 diagnostic codes. Data were extracted on demographics, clinical presentation, fracture characteristics, and details of treatment. RESULTS A total of 383 patients (median age 78 years, 75.2% male) were included. Over a median follow-up of 17.3 months, overall mortality was 38.1%, in-hospital mortality was 5.2%, 30-day mortality was 5.7%, and 1-year mortality was 18.0%. Initial treatment consisted of external orthosis in 220 (57.4%) and surgery in 144 (37.6%), while 19 (5.0%) were left untreated. Patients who died were older (median 83 vs 74 years, p < 0.001), had a lower BMI (29.4 vs 32.3 kg/m 2 , p < 0.001), had higher frailty on the 5-item modified frailty index (mFI-5), and more commonly had congestive heart failure (CHF), cancer, functional dependency, and chronic kidney disease (CKD) (all p < 0.01). Mortality was not associated with fracture location, mechanism of injury, fracture number, or the presence of neurological deficits. Univariable time-to-event analysis showed that decreased survival was predicted by older age (HR 1.08, p < 0.0001), lower BMI (HR per kg/m 2 0.93, p < 0.0001), functional dependency (HR 2.05, p < 0.001), cancer (HR 1.55 p = 0.01), CKD (HR 1.48, p = 0.019), CHF (HR 2.18, p < 0.001), and higher mFI-5 score (HR per point 1.34, p < 0.001). On multivariable analysis, older age (HR 1.06 per year, p < 0.001), lower BMI (HR per kg/m 2 0.95, p < 0.001), and mFI-5 score (HR per patient 1.37, p < 0.001) remained significant independent predictors of mortality. Treatment modality did not predict mortality. CONCLUSIONS AS- or DISH-associated spinal fractures have a high associated mortality, regardless of mechanism or severity. Treatment modality did not affect odds of survival, suggesting that nonoperative management with an external orthosis may be an appropriate option, particularly in older and medically complex patients.
Operationalizing WHO Ethical Principles for Healthcare AI: A Lifecycle-Aligned Governance-by-Design Framework
Saraboji, Kaaviyashri , Gopalakrishnan, Keerthy, Sood, Divyanshi, Kaur, Anmolpreet , Shivaram, Suganti, Helgeson, Scott A. , Arunachalam, Shivaram P. , Mitra, Dipankar , Kaur, Anmolpreet , Arunachalam, S P , Helgeson, Scott A

Mayo Clinic

AI in Medicine | June 10, 2026 | Volume: 1 | Issue: 2 | Pages: 16

10.3390/aimed1020016 Get It

Artificial intelligence (AI) is rapidly transforming healthcare through applications in clinical decision support, diagnostic imaging, population health management, and workflow optimization. Despite these advances, real-world deployment continues to expose critical challenges related to safety, bias, transparency, and integration into clinical workflows. Algorithmic bias can exacerbate health disparities, limited explainability may undermine clinician trust, and insufficient validation and post-deployment monitoring can compromise patient safety. Although the World Health Organization (WHO) has established six ethical principles for AI in health, including autonomy, well-being and safety, transparency, accountability, equity, and sustainability, translating these high-level principles into practical and enforceable governance mechanisms remains a persistent challenge. This narrative review synthesizes insights from bioethics, health policy, computer science, and clinical medicine to identify gaps in current AI governance approaches and proposes a lifecycle-aligned governance-by-design framework that operationalizes WHO ethical principles across key stages of the healthcare AI lifecycle, including data collection, model development, validation, deployment, and post-deployment monitoring. The framework integrates concrete governance mechanisms such as consent governance, fairness evaluation, external validation, explainability, clinician oversight, and continuous performance monitoring. Overall, this work advances a practical, lifecycle-integrated approach to AI governance and provides a structured foundation for developing safe, equitable, and trustworthy AI systems in healthcare.
A Novel AI Powered Objective Pain Quantifying Software as a Medical Device
Shariff, Mohammad Naveed, LEE, JIEUN , Yerrapragada, Gayathri, Elangovan, Poonguzhali, Janarthanan, Jayarajasekaran, Richard, Samuel, Mitra, Dipankar , Sood, Divyanshi, Iyer, Vivek N., Arunachalam, S., Arunachalam, S P , Shariff, Mohammed Naveed , Yerrapragada, Gayathri, Elangovan, Poonguzhali , Natarajan, Thangeswaran, Helgeson, Scott A. , Arunachalam, Shivaram P.

Mayo Clinic

2026 Design of Medical Devices Conference | June 4, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.1115/dmd2026-1045 Get It

Abstract Pain assessment plays a critical role in clinical decision-making, yet conventional methods rely heavily on subjective self-reports such as the Visual Analog Scale (VAS) or the Numeric Rating Scale (NRS). These approaches fail in scenarios involving infants, sedated patients, unconscious individuals, or those with cognitive or communication impairments. To address this gap, this research proposes PainNet, an AI-driven, real-time facial pain recognition system that integrates deep learning-based object detection with Action Unit (AU)–based pain intensity scoring. The system uses YOLOv8 for rapid facial expression detection and incorporates Py-Feat to extract pain-relevant AUs, enabling computation of the Prkachin and Solomon Pain Intensity (PSPI) score. A balanced dataset comprising 500 images sourced from Delaware, BioVid, and RaFD was curated, annotated, and preprocessed for training. The YOLOv8-based binary pain vs. no-pain classifier achieved a mean Average Precision (mAP) of 0.89 and an F1-score of 0.80 on the test dataset, demonstrating strong performance in diverse environments and facial variations. The design emphasizes computational efficiency by activating AU analysis only when YOLOv8 detects probable pain, making PainNet suitable for continuous monitoring in ICUs, elder-care facilities, and telemedicine systems. This study demonstrates that the fusion of real-time object detection and clinically validated AU scoring can significantly advance automated, objective, and scalable pain assessment.
A Novel you only Listen once (YOLO) Deep Learning Model for Interstitial Lung Disease (ILD) Detection using Digital Lung Sound Auscultation: A Feasibility Study
LEE, JIEUN , Yerrapragada, Gayathri, Shariff, Mohammad Naveed, Ibrahim, Mohamed I. , Khalifa, H. Ezzat , Handral, Megha Sasidhar, Cherukuri, Swathi Priya , Kaur, Avneet , Sood, Divyanshi, IYER, VIVEK N, Arunachalam, S., Arunachalam, S P , Yerrapragada, Gayathri, Shariff, Mohammed Naveed , Helgeson, Scott A. , Arunachalam, Shivaram P.

Mayo Clinic

2026 Design of Medical Devices Conference | June 4, 2026 | Volume: No volume available | Issue: No issue available | Pages: No page available

10.1115/dmd2026-1043 Get It

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.