Proceedings of SPIE--the International Society for Optical Engineering最新文献

{"title":"AI-Driven Multi-Lesion Detection in Whole-body FDG PET/CT.","authors":"Xiaofeng Liu, Menghua Xia, Yanis Chemli, Georges El Fakhri, Chi Liu, Jinsong Ouyang","doi":"10.1117/12.3087729","DOIUrl":"10.1117/12.3087729","url":null,"abstract":"<p><p>An automated multi-lesion detection tool for whole-body FDG PET/CT can reduce radiologist workload and improve diagnostic accuracy by consistently identifying subtle or numerous lesions, enabling faster workflows and better treatment planning for patients with widespread disease. In this work, we investigate lesion detectability and localizability with a deep learning model, comparing the performance using PET and CT alone versus combined PET+CT data. We leverage a large public dataset of 1,014 FDG-PET/CT studies with lesion annotations, converted to 3D bounding boxes for detection training. A 3D object detector (RetinaNet) using the MONAI framework was learned on 700 training studies and 114 validation studies, optimizing a custom anchor configuration derived from clustering of lesion sizes. The detector was evaluated on 200 independent test studies for lesion detection and localization performance. In addition to conventional average precision (AP) at various intersection-over-union (IoU) thresholds, free-response ROC (FROC) analysis was also used to quantify lesion detectability and localizability. The proposed method achieved robust lesion detection and localization, with the PET+CT model outperforming PET-only across all IoU thresholds and better FROC. The results indicate that the incorporation of CT anatomical information significantly improves the detectability of lesions. This study highlights a detection paradigm for whole-body PET/CT lesion analysis, showing improved performance with multimodal PET+CT input. The findings suggest that 3D detection models can effectively identify and localize lesions and benefit from CT information, supporting their potential use in computer-aided detection for oncologic PET/CT. These results demonstrate that AI-based lesion detection networks are approaching the level of performance that may support future clinical deployment, with the potential to greatly enhance the efficiency of radiologists or even fully automate certain detection tasks.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13928 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13671731/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149217378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"How Close Are We? Limitations and Progress of AI Models in Banff Lesion Scoring.","authors":"Yanfan Zhu, Juming Xiong, Ruining Deng, Yu Wang, Yaohong Wang, Shilin Zhao, Mengmeng Yin, Yuqing Liu, Haichun Yang, Yuankai Huo","doi":"10.1117/12.3086220","DOIUrl":"10.1117/12.3086220","url":null,"abstract":"<p><p>The Banff Classification provides the global standard for evaluating renal transplant biopsies, yet its semi-quantitative nature, complex criteria, and inter-observer variability present significant challenges for computational replication. In this study, we explore the feasibility of approximating Banff lesion scores using existing deep learning models through a modular, rule-based framework. We decompose each Banff indicator-such as glomerulitis <math><mo>(</mo> <mi>g</mi> <mo>)</mo></math> , peritubular capillaritis <math><mo>(</mo> <mi>p</mi> <mi>t</mi> <mi>c</mi> <mo>)</mo></math> , and intimal arteritis <math><mo>(</mo> <mi>v</mi> <mo>)</mo></math> -into its constituent structural and inflammatory components, and assess whether current segmentation and detection tools can support their computation. Model outputs are mapped to Banff scores using heuristic rules aligned with expert guidelines, and evaluated against expert-annotated ground truths. Our findings highlight both partial successes and critical failure modes, including structural omission, hallucination, and detection ambiguity. Even when final scores match expert annotations, inconsistencies in intermediate representations often undermine interpretability. These results reveal the limitations of current AI pipelines in replicating computational expert-level grading, and emphasize the importance of modular evaluation and computational Banff grading standard in guiding future model development for transplant pathology.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13932 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13591582/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149019123","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MedFoundationHub: A Lightweight and Secure Toolkit for Deploying Medical Vision Language Foundation Models.","authors":"Xiao Li, Yanfan Zhu, Ruining Deng, Wei-Qi Wei, Yu Wang, Shilin Zhao, Yaohong Wang, Haichun Yang, Yuankai Huo","doi":"10.1117/12.3087828","DOIUrl":"10.1117/12.3087828","url":null,"abstract":"<p><p>Recent advances in medical vision-language models (VLMs) open up remarkable opportunities for clinical applications such as automated report generation, physician copilots, and uncertainty quantification. Despite their promise, medical VLMs raise serious security concerns. These include the risk of Protected Health Information (PHI) exposure, data leakage, and vulnerability to cyberthreats, concerns that are especially critical in hospital environments. Even when adopted for research or non-clinical purposes, healthcare organizations must exercise caution and implement safeguards. To address these challenges, we present MedFoundationHub, a graphical user interface (GUI) toolkit that: (1) enables physicians to manually select and use different models without programming expertise, (2) supports engineers in efficiently deploying medical VLMs in a plug-and-play fashion, with seamless integration of Hugging Face open-source models, and (3) ensures privacy-preserving inference through Docker-orchestrated, operating system agnostic deployment. MedFoundationHub requires only an offline local workstation equipped with a single NVIDIA A6000 GPU, making it both secure and accessible within the typical resources of academic research labs. To evaluate current capabilities, we engaged board-certified pathologists to deploy and assess five state-of-the-art VLMs (Google-MedGemma3-4B, Qwen2-VL-7B-Instruct, Qwen2.5-VL-7B-Instruct, and LLaVA-1.5-7B/13B). Expert evaluation covered colon cases and renal cases, yielding 1,015 clinician-model scoring events. These assessments revealed recurring limitations, including off-target answers, vague reasoning, and inconsistent pathology terminology.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13928 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13590015/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149013340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatial frequency domain assessment of burns in a clinical setting: a comparison of a research-grade and a commercial imager at the UCI Regional Burn Center.","authors":"Gordon T Kennedy, Christopher A Campbell, Alexis M Fox, Robert J Christy, Theresa Chin, Victor Joe, Anthony J Durkin","doi":"10.1117/12.3081218","DOIUrl":"https://doi.org/10.1117/12.3081218","url":null,"abstract":"<p><p>Accurate assessment of burn wounds is critical for determining optimal patient care. The current standard of bedside visual and tactile examination is subjective and has poor accuracy, even when performed by skilled clinicians. Imaging technologies offer more accurate burn classification and have the potential to be used by personnel with less burn-centric training. This study evaluates the performance of two spatial frequency domain imaging (SFDI) systems in a burn intensive care unit (ICU) to classify burn wounds and identify key measurement parameters for future device development. Subjects (N=24) at the University of California, Irvine (UCI) Regional Burn Center were imaged under an IRB protocol using the OxImager RS (research device) and Clarifi RS (clinical device). We present side-by-side comparisons of device performance metrics, including optical property maps across a range of burn severities, to determine optimal SFDI parameters for clinical assessment. Results indicate that while the research-grade system provides a higher resolution of spectral and spatial data, the streamlined clinical system offers significant advantages in acquisition speed and resistance to motion artifacts and ambient light in the ICU environment.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13823 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13652301/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149167616","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A standardized data specification for reproducible, verifiable structured light and multispectral image sets.","authors":"Christopher A Campbell, Pratik Nudurupati, Anthony J Durkin","doi":"10.1117/12.3081186","DOIUrl":"https://doi.org/10.1117/12.3081186","url":null,"abstract":"<p><p>Spatial Frequency Domain Imaging (SFDI) is a quantitative wide-field imaging technique that uses light and diffuse optical spectroscopic principles to rapidly generate tissue optical property images. However, SFDI datasets present significant workflow challenges: they can be large (on the order of gigabytes), contain multiple image formats, and require computationally intensive processing. Additionally, SFDI images typically require calibration measurements, further complicating data organization. The proliferation of calibration data, imaging data, and processing options creates technical overhead and leads to reproducibility issues. While multiple \"Open SFDI\" initiatives exist, no standardized storage format has emerged, making the replication of published SFDI research difficult. Commercial solutions like Modulim's directory-based format contain trade secrets and lack open-source processing options. We are developing an open standard for SFDI images, extensible to multispectral images, dubbed SpectralMaps. This specification leverages modern Xarray data structures saved in Zarr format, offering fast loading of complete datasets or selected data chunks. Zarr permits lossless compression with partial loading of data, making data loading faster and partial data processing possible without loading the entire dataset. Metadata for measurements and processing history are included in the file. Datasets are identified by hashes included for verification. This specification provides a unified, transparent framework for storing, sharing, and processing SFDI data, reducing technical overhead and improving reproducibility across research groups. By adopting an open, extensible standard, the community can accelerate method development, enable cross-platform interoperability, and foster collaborative validation of SFDI techniques.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13840 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13652299/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149169360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hybrid-Intelligence Framework for Pathological Liver Segmentation and Disease Localization in CT Imaging.","authors":"Deonte Hall, Jayaram K Udupa, Yadavendra Nln, Yubing Tong, Caiyun Wu, Drew A Torigian","doi":"10.1117/12.3087377","DOIUrl":"10.1117/12.3087377","url":null,"abstract":"","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13926 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13676773/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149221249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparing evaluation of quantitative imaging methods using reference standards versus regression‑without‑truth‑based technique.","authors":"Yan Liu, Abhinav K Jha","doi":"10.1117/12.3087777","DOIUrl":"10.1117/12.3087777","url":null,"abstract":"<p><p>Clinical translation of quantitative imaging (QI) methods requires objective evaluation of these methods on reliably measuring the underlying true quantitative values. Ideally, such evaluation would be performed using ground truth or gold standards. However, in clinical practice, ground truth is generally unavailable, and obtaining gold standards for large patient cohorts is often impractical or infeasible. In these scenarios, evaluation using well-accepted or commonly used methods for measuring the quantitative values that may have bias and/or measurement error, referred to as reference standards, is a common alternative. However, the error in reference standards might lead to inaccurate results in evaluating the QI methods. To address the challenge of evaluation without a gold standard, a class of regression-without-truth (RWT)-based techniques has also been proposed. These techniques compare QI methods without requiring ground truth and have demonstrated promise across multiple applications. This raises an important question: how does evaluation using reference standards vs. RWT-based techniques compare? Our work, which investigates this question using controlled numerical experiments, showed that when the measurement error of the reference standard exceeded a certain level, evaluation with a RWT-based technique outperformed evaluation with reference standards. This study provides insights for selecting appropriate evaluation strategies when gold standards are unavailable and motivates further investigation in clinically realistic settings.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13928 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13483360/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148802374","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Experimental Validation and Calibration of Cherenkov Imaging for Real-Time Surface Dosimetry in Photon Breast Therapy.","authors":"Baozhu Lu, Yifeng Zhu, Lisha Chen, Karen Teng, Emily Xiong, Taoran Li, Gary Freedman, Neil Taunk, Brian W Pogue, Timothy C Zhu","doi":"10.1117/12.3081551","DOIUrl":"10.1117/12.3081551","url":null,"abstract":"<p><p>Cherenkov imaging holds promise as a non-invasive, real-time method for surface dosimetry during photon-based external beam radiation therapy (EBRT). In this study, we evaluated the feasibility of using Cherenkov emission as a quantitative surrogate for surface dose in breast cancer treatment through Monte Carlo simulation and experimental validation. A Monte Carlo model was developed to simulate both dose deposition and corresponding Cherenkov photon emission for breast patients undergoing photon EBRT. Validation was performed using phantoms with homogeneous optical properties and varying geometries, enabling direct comparison between simulated and measured Cherenkov emission. Additionally, the simulated surface dose was validated against in-vivo measurements acquired using in-vivo dosimeters. Results show strong correlation between simulated and measured Cherenkov emission and surface dose, except at the field edges, where additional optical and geometric considerations are required. Based on these comparisons, a correction factor was derived to calibrate Cherenkov intensity to absolute surface dose. Finally, Cherenkov images acquired from breast cancer patients were compared with simulation results to assess the clinical utility of this approach. These findings support the use of Cherenkov imaging, when properly corrected, as a viable tool for real-time, patient-specific surface dosimetry in photon EBRT.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13841 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13505574/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148820723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Automated Classification of Skeletal Facial Asymmetry in CBCT Using a Reproducible 3D Slicer Workflow for Patient-Specific Decision Support.","authors":"Angelina Sanna, Janson Ng, Antonio Ruellas, Aron Aliaga, Alban Gaydamour, Enzo Tulissi, Yanjie Huang, Juan Carlos Prieto, Najla Al Turkestani, Marilla Yatabe, Jonas Bianchi, Steve Pieper, Lucia Cevidanes, Tengfei Li","doi":"10.1117/12.3087010","DOIUrl":"https://doi.org/10.1117/12.3087010","url":null,"abstract":"<p><p>This study presents an automated pipeline for classifying skeletal facial asymmetry of 170 patients with Class III malocclusion using cone-beam computed tomography (CBCT). The methodology integrates standardized image preprocessing, mirroring, voxel-based registration, anatomical measurement extraction, dimensionality reduction, and clustering. Principal component analysis (PCA), t-distributed stochastic neighbor embedding (t-SNE), and uniform manifold approximation and projection (UMAP) were evaluated using Spectral Clustering with both k-nearest-neighbors (kNN) and radial basis function (RBF) affinities. PCA suggested four clusters, t-SNE found four with kNN and two with RBF, and UMAP identified six. UMAP with RBF affinity achieved the best overall clustering performance, as indicated by the silhouette score, C-index, Dunn index, and closed surface distance heat maps, slightly outperforming UMAP with kNN. The unsupervised framework established in this study provides clinically meaningful phenotypes that could inform subsequent supervised classification work with an expanded dataset. The pipeline is implemented within open-source 3D Slicer Craniomaxillofacial automated modules to enhance reproducibility and clinical translation.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13929 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13630422/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149119715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Composite CT scoring system for Diffuse Idiopathic Skeletal Hyperostosis (DISH): a validated and reproducible framework for diagnosis and severity assessment.","authors":"Lily Gao, Andrew Hoyek, Enobong Anako, Kathryn Jordan, Trey McGonigle, Shi Huang, Manish Kota, Bennett Landman, Katherine Van Schaik","doi":"10.1117/12.3077719","DOIUrl":"10.1117/12.3077719","url":null,"abstract":"<p><p>Variability among observers remains a major barrier to standardizing imaging findings for spinal disease, hindering clinical diagnosis, research reproducibility, and the development of automated tools for assessment of spinal disease. Diffuse Idiopathic Skeletal Hyperostosis (DISH) is underdiagnosed in part due to the lack of consistent, CT-adapted scoring frameworks. We develop and validate a reproducible CT-based scoring system that integrates level-wise ossification scoring (Kuperus) with a structured severity classification (Holton) to capture DISH presence and bridging severity based on the number and distribution of bridged levels (scored 2 or 3). Five raters of varying experience levels (undergraduate students to an attending radiologist) scored 24 training scans; junior raters resolved discrepancies collaboratively, while senior raters reviewed independently. Pairwise comparisons showed perfect specificity (1.00) and moderate sensitivity (0.62-0.75) for DISH presence among less experienced raters compared to the attending radiologist. McNemar's tests showed no significant differences in pairwise disagreement rates for DISH presence, while paired Wilcoxon tests revealed significant differences in bridged level counts, with less experienced raters underestimating the number of bridged levels. Scoring time was 2-3 minutes per scan following training. Subsequently, two independent raters scored 614 thoracolumbar CT scans. Interobserver reliability was excellent for DISH presence (Gwet's AC2 = 0.957 [95% CI: 0.936-0.977]), bridged level count (ICC = 0.933 [95% CI: 0.922-0.943]), and bridging severity (AC2 = 0.950 [95% CI: 0.931-0.969]). Level-wise scores also showed high agreement (median AC2 = 0.881, range: 0.862-0.937), with discrepancies occurring primarily due to transitional anatomy and mildly ossified levels. This CT-based scoring framework enables reliable and reproducible labeling of DISH presence and severity across raters of varying experience levels. Its structured format supports scalable image annotation and minimizes label noise, providing a robust foundation for training computer vision models to automate detection and stratification of DISH in spinal imaging.</p>","PeriodicalId":74505,"journal":{"name":"Proceedings of SPIE--the International Society for Optical Engineering","volume":"13928 ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13348741/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148427097","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}