Ovarian Loss in Laparoscopic and Robotic Cystectomy Compared Using Artificial Intelligence Pathology.

A Comparison of Ovarian Loss Following Laparoscopic versus Robotic Cystectomy As Analyzed by Artificial Intelligence-Powered Pathology Software. Background and Objective: To compare the area of ovarian tissue and follicular loss in the excised cystectomy specimen of endometrioma performed by laparoscopic or robotic technique. Methods: Prospective observational study performed between April 2023 to August 2023. There were 14 patients each in Laparoscopic group (LC) and Robotic group (RC). Excised cyst wall sent was for to the pathologist who was blinded to the technique used for cystectomy. The pathological assessment was done by artificial intelligence-Whole Slide Imaging (WSI) software. Results: The age was significantly lower in LC group; the rest of demographic results were comparable. The mean of the median ovarian area loss [Mean Rank, LC group (9.1 ± 15.1); RC (8.1 ± 12.4)] was higher in LC group. The mean of the median total follicular loss was higher in LC group (8.9 ± 9.2) when compared to RC group (6.3 ± 8.9) and was not significant. The area of ovarian loss in bilateral endometrioma was significantly higher in LC group (mean rank 7.5) as compared to RC group (mean rank 3) - (P = .016) despite more cases of bilateral disease in RC group. With increasing cyst size the LC group showed increased median loss of follicles when compared to RC group (strong correlation coefficient 0.347) but not statistically significant (P = .225). AAGL (American Association of Gynecologic Laparoscopists) score did not have any impact on the two techniques. Conclusion: Robotic assistance reduces the area of ovarian and follicular loss during cystectomy of endometrioma especially in bilateral disease and increasing cyst size. It should be considered over the laparoscopic approach if available.

Quality of Large Language Model Responses to Radiation Oncology Patient Care Questions.

Importance: Artificial intelligence (AI) large language models (LLMs) demonstrate potential in simulating human-like dialogue. Their efficacy in accurate patient-clinician communication within radiation oncology has yet to be explored. Objective: To determine an LLM's quality of responses to radiation oncology patient care questions using both domain-specific expertise and domain-agnostic metrics. Design, Setting, and Participants: This cross-sectional study retrieved questions and answers from websites (accessed February 1 to March 20, 2023) affiliated with the National Cancer Institute and the Radiological Society of North America. These questions were used as queries for an AI LLM, ChatGPT version 3.5 (accessed February 20 to April 20, 2023), to prompt LLM-generated responses. Three radiation oncologists and 3 radiation physicists ranked the LLM-generated responses for relative factual correctness, relative completeness, and relative conciseness compared with online expert answers. Statistical analysis was performed from July to October 2023. Main Outcomes and Measures: The LLM's responses were ranked by experts using domain-specific metrics such as relative correctness, conciseness, completeness, and potential harm compared with online expert answers on a 5-point Likert scale. Domain-agnostic metrics encompassing cosine similarity scores, readability scores, word count, lexicon, and syllable counts were computed as independent quality checks for LLM-generated responses. Results: Of the 115 radiation oncology questions retrieved from 4 professional society websites, the LLM performed the same or better in 108 responses (94%) for relative correctness, 89 responses (77%) for completeness, and 105 responses (91%) for conciseness compared with expert answers. Only 2 LLM responses were ranked as having potential harm. The mean (SD) readability consensus score for expert answers was 10.63 (3.17) vs 13.64 (2.22) for LLM answers (P < .001), indicating 10th grade and college reading levels, respectively. The mean (SD) number of syllables was 327.35 (277.15) for expert vs 376.21 (107.89) for LLM answers (P = .07), the mean (SD) word count was 226.33 (191.92) for expert vs 246.26 (69.36) for LLM answers (P = .27), and the mean (SD) lexicon score was 200.15 (171.28) for expert vs 219.10 (61.59) for LLM answers (P = .24). Conclusions and Relevance: In this cross-sectional study, the LLM generated accurate, comprehensive, and concise responses with minimal risk of harm, using language similar to human experts but at a higher reading level. These findings suggest the LLM's potential, with some retraining, as a valuable resource for patient queries in radiation oncology and other medical fields.

Utilizing geospatial artificial intelligence to map cancer disparities across health regions.

We have developed an innovative tool, the Intelligent Catchment Analysis Tool (iCAT), designed to identify and address healthcare disparities across specific regions. Powered by Artificial Intelligence and Machine Learning, our tool employs a robust Geographic Information System (GIS) to map healthcare outcomes and disease disparities. iCAT allows users to query publicly available data sources, health system data, and treatment data, offering insights into gaps and disparities in diagnosis and treatment paradigms. This project aims to promote best practices to bridge the gap in healthcare access, resources, education, and economic opportunities. The project aims to engage local and regional stakeholders in data collection and evaluation, including patients, providers, and organizations. Their active involvement helps refine the platform and guides targeted interventions for more effective outcomes. In this paper, we present two sample illustrations demonstrating how iCAT identifies healthcare disparities and analyzes the impact of social and environmental variables on outcomes. Over time, this platform can help communities make decisions to optimize resource allocation.

Accelerating Drug Development Using Spatial Multi-omics.

SUMMARY: Spatial biology approaches enabled by innovations in imaging biomarker platforms and artificial intelligence-enabled data integration and analysis provide an assessment of patient and disease heterogeneity at ever-increasing resolution. The utility of spatial biology data in accelerating drug programs, however, requires balancing exploratory discovery investigations against scalable and clinically applicable spatial biomarker analysis.

The 3D Revolution in Cancer Discovery.

SUMMARY: The transition from 2D to 3D spatial profiling marks a revolutionary era in cancer research, offering unprecedented potential to enhance cancer diagnosis and treatment. This commentary outlines the experimental and computational advancements and challenges in 3D spatial molecular profiling, underscoring the innovation needed in imaging tools, software, artificial intelligence, and machine learning to overcome implementation hurdles and harness the full potential of 3D analysis in the field.

Radiomic Application for Head and Neck Squamocellular Tumor: Systematic Review.

Abstract: Radiomics represents the convergence of artificial intelligence and radiological data analysis, primarily applied in the diagnosis and treatment of cancer. In the head and neck region, squamous cell carcinoma is the most prevalent type of tumor. Recent radiomics research has revealed that specific bio-imaging characteristics correlate with various molecular features of Head and Neck Squamous Cell Carcinoma (HNSCC), particularly Human Papillomavirus (HPV). These tumors typically present a unique phenotype, often affecting younger patients, and show a favorable response to radiation therapy. This study provides a systematic review of the literature, summarizing the application of radiomics in the head and neck region. It offers a comprehensive analysis of radiomics-based studies on HNSCC, evaluating its potential for tumor evaluation, risk stratification, and outcome prediction in head and neck cancer treatment.

[Artificial intelligence in pathological anatomy]. / Iskusstvennyi intellekt v patologicheskoi anatomii.

The review presents key concepts and global developments in the field of artificial intelligence used in pathological anatomy. The work examines two types of artificial intelligence (AI): weak and strong ones. A review of experimental algorithms using both deep machine learning and computer vision technologies to work with WSI images of preparations, diagnose and make a prognosis for various malignant neoplasms is carried out. It has been established that weak artificial intelligence at this stage of development of computer (digital) pathological anatomy shows significantly better results in speeding up and refining diagnostic procedures than strong artificial intelligence having signs of general intelligence. The article also discusses three options for the further development of AI assistants for pathologists based on the technologies of large language models (strong AI) ChatGPT (PathAsst), Flan-PaLM2 and LIMA. As a result of the analysis of the literature, key problems in the field were identified: the equipment of pathology institutions, the lack of experts in training neural networks, the lack of strict criteria for the clinical viability of AI diagnostic technologies.

Use of an AI Score Combining Cancer Signs, Masking, and Risk to Select Patients for Supplemental Breast Cancer Screening.

Background Mammographic density measurements are used to identify patients who should undergo supplemental imaging for breast cancer detection, but artificial intelligence (AI) image analysis may be more effective. Purpose To assess whether AISmartDensity-an AI-based score integrating cancer signs, masking, and risk-surpasses measurements of mammographic density in identifying patients for supplemental breast imaging after a negative screening mammogram. Materials and Methods This retrospective study included randomly selected individuals who underwent screening mammography at Karolinska University Hospital between January 2008 and December 2015. The models in AISmartDensity were trained and validated using nonoverlapping data. The ability of AISmartDensity to identify future cancer in patients with a negative screening mammogram was evaluated and compared with that of mammographic density models. Sensitivity and positive predictive value (PPV) were calculated for the top 8% of scores, mimicking the proportion of patients in the Breast Imaging Reporting and Data System "extremely dense" category. Model performance was evaluated using area under the receiver operating characteristic curve (AUC) and was compared using the DeLong test. Results The study population included 65 325 examinations (median patient age, 53 years [IQR, 47-62 years])-64 870 examinations in healthy patients and 455 examinations in patients with breast cancer diagnosed within 3 years of a negative screening mammogram. The AUC for detecting subsequent cancers was 0.72 and 0.61 (P < .001) for AISmartDensity and the best-performing density model (age-adjusted dense area), respectively. For examinations with scores in the top 8%, AISmartDensity identified 152 of 455 (33%) future cancers with a PPV of 2.91%, whereas the best-performing density model (age-adjusted dense area) identified 57 of 455 (13%) future cancers with a PPV of 1.09% (P < .001). AISmartDensity identified 32% (41 of 130) and 34% (111 of 325) of interval and next-round screen-detected cancers, whereas the best-performing density model (dense area) identified 16% (21 of 130) and 9% (30 of 325), respectively. Conclusion AISmartDensity, integrating cancer signs, masking, and risk, outperformed traditional density models in identifying patients for supplemental imaging after a negative screening mammogram. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Kim and Chang in this issue.

Tertiary lymphoid structural heterogeneity determines tumour immunity and prospects for clinical application.

Tertiary lymphoid structures (TLS) are clusters of immune cells that resemble and function similarly to secondary lymphoid organs (SLOs). While TLS is generally associated with an anti-tumour immune response in most cancer types, it has also been observed to act as a pro-tumour immune response. The heterogeneity of TLS function is largely determined by the composition of tumour-infiltrating lymphocytes (TILs) and the balance of cell subsets within the tumour-associated TLS (TA-TLS). TA-TLS of varying maturity, density, and location may have opposing effects on tumour immunity. Higher maturity and/or higher density TLS are often associated with favorable clinical outcomes and immunotherapeutic response, mainly due to crosstalk between different proportions of immune cell subpopulations in TA-TLS. Therefore, TLS can be used as a marker to predict the efficacy of immunotherapy in immune checkpoint blockade (ICB). Developing efficient imaging and induction methods to study TA-TLS is crucial for enhancing anti-tumour immunity. The integration of imaging techniques with biological materials, including nanoprobes and hydrogels, alongside artificial intelligence (AI), enables non-invasive in vivo visualization of TLS. In this review, we explore the dynamic interactions among T and B cell subpopulations of varying phenotypes that contribute to the structural and functional diversity of TLS, examining both existing and emerging techniques for TLS imaging and induction, focusing on cancer immunotherapies and biomaterials. We also highlight novel therapeutic approaches of TLS that are being explored with the aim of increasing ICB treatment efficacy and predicting prognosis.

Application of the convolution neural network in determining the depth of invasion of gastrointestinal cancer: a systematic review and meta-analysis.

BACKGROUND: With the development of endoscopic technology, endoscopic submucosal dissection (ESD) has been widely used in the treatment of gastrointestinal tumors. It is necessary to evaluate the depth of tumor invasion before the application of ESD. The convolution neural network (CNN) is a type of artificial intelligence that has the potential to assist in the classification of the depth of invasion in endoscopic images. This meta-analysis aimed to evaluate the performance of CNN in determining the depth of invasion of gastrointestinal tumors. METHODS: A search on PubMed, Web of Science, and SinoMed was performed to collect the original publications about the use of CNN in determining the depth of invasion of gastrointestinal neoplasms. Pooled sensitivity and specificity were calculated using an exact binominal rendition of the bivariate mixed-effects regression model. I2 was used for the evaluation of heterogeneity. RESULTS: A total of 17 articles were included; the pooled sensitivity was 84% (95% CI, 0.81-0.88), specificity was 91% (95% CI, 0.85-0.94), and the area under the curve (AUC) was 0.93 (95% CI, 0.90-0.95). The performance of CNN was significantly better than that of endoscopists (AUC: 0.93 vs 0.83, respectively; P = .0005). CONCLUSION: Our review revealed that CNN is one of the most effective methods of endoscopy to evaluate the depth of invasion of early gastrointestinal tumors, which has the potential to work as a remarkable tool for clinical endoscopists to make decisions on whether the lesion is feasible for endoscopic treatment.

Artificial intelligence large language model ChatGPT: is it a trustworthy and reliable source of information for sarcoma patients?

Introduction: Since its introduction in November 2022, the artificial intelligence large language model ChatGPT has taken the world by storm. Among other applications it can be used by patients as a source of information on diseases and their treatments. However, little is known about the quality of the sarcoma-related information ChatGPT provides. We therefore aimed at analyzing how sarcoma experts evaluate the quality of ChatGPT's responses on sarcoma-related inquiries and assess the bot's answers in specific evaluation metrics. Methods: The ChatGPT responses to a sample of 25 sarcoma-related questions (5 definitions, 9 general questions, and 11 treatment-related inquiries) were evaluated by 3 independent sarcoma experts. Each response was compared with authoritative resources and international guidelines and graded on 5 different metrics using a 5-point Likert scale: completeness, misleadingness, accuracy, being up-to-date, and appropriateness. This resulted in maximum 25 and minimum 5 points per answer, with higher scores indicating a higher response quality. Scores ≥21 points were rated as very good, between 16 and 20 as good, while scores ≤15 points were classified as poor (11-15) and very poor (≤10). Results: The median score that ChatGPT's answers achieved was 18.3 points (IQR, i.e., Inter-Quartile Range, 12.3-20.3 points). Six answers were classified as very good, 9 as good, while 5 answers each were rated as poor and very poor. The best scores were documented in the evaluation of how appropriate the response was for patients (median, 3.7 points; IQR, 2.5-4.2 points), which were significantly higher compared to the accuracy scores (median, 3.3 points; IQR, 2.0-4.2 points; p = 0.035). ChatGPT fared considerably worse with treatment-related questions, with only 45% of its responses classified as good or very good, compared to general questions (78% of responses good/very good) and definitions (60% of responses good/very good). Discussion: The answers ChatGPT provided on a rare disease, such as sarcoma, were found to be of very inconsistent quality, with some answers being classified as very good and others as very poor. Sarcoma physicians should be aware of the risks of misinformation that ChatGPT poses and advise their patients accordingly.

Generative artificial intelligence and non-pharmacological bias: an experimental study on cancer patient sexual health communications.

Objectives The objective of this study was to explore the feature of generative artificial intelligence (AI) in asking sexual health among cancer survivors, which are often challenging for patients to discuss.Methods We employed the Generative Pre-trained Transformer-3.5 (GPT) as the generative AI platform and used DocsBot for citation retrieval (June 2023). A structured prompt was devised to generate 100 questions from the AI, based on epidemiological survey data regarding sexual difficulties among cancer survivors. These questions were submitted to Bot1 (standard GPT) and Bot2 (sourced from two clinical guidelines).Results No censorship of sexual expressions or medical terms occurred. Despite the lack of reflection on guideline recommendations, 'consultation' was significantly more prevalent in both bots' responses compared with pharmacological interventions, with ORs of 47.3 (p<0.001) in Bot1 and 97.2 (p<0.001) in Bot2.Discussion Generative AI can serve to provide health information on sensitive topics such as sexual health, despite the potential for policy-restricted content. Responses were biased towards non-pharmacological interventions, which is probably due to a GPT model designed with the 's prohibition policy on replying to medical topics. This shift warrants attention as it could potentially trigger patients' expectations for non-pharmacological interventions.

Public data homogenization for AI model development in breast cancer.

BACKGROUND: Developing trustworthy artificial intelligence (AI) models for clinical applications requires access to clinical and imaging data cohorts. Reusing of publicly available datasets has the potential to fill this gap. Specifically in the domain of breast cancer, a large archive of publicly accessible medical images along with the corresponding clinical data is available at The Cancer Imaging Archive (TCIA). However, existing datasets cannot be directly used as they are heterogeneous and cannot be effectively filtered for selecting specific image types required to develop AI models. This work focuses on the development of a homogenized dataset in the domain of breast cancer including clinical and imaging data. METHODS: Five datasets were acquired from the TCIA and were harmonized. For the clinical data harmonization, a common data model was developed and a repeatable, documented "extract-transform-load" process was defined and executed for their homogenization. Further, Digital Imaging and COmmunications in Medicine (DICOM) information was extracted from magnetic resonance imaging (MRI) data and made accessible and searchable. RESULTS: The resulting harmonized dataset includes information about 2,035 subjects with breast cancer. Further, a platform named RV-Cherry-Picker enables search over both the clinical and diagnostic imaging datasets, providing unified access, facilitating the downloading of all study imaging that correspond to specific series' characteristics (e.g., dynamic contrast-enhanced series), and reducing the burden of acquiring the appropriate set of images for the respective AI model scenario. CONCLUSIONS: RV-Cherry-Picker provides access to the largest, publicly available, homogenized, imaging/clinical dataset for breast cancer to develop AI models on top. RELEVANCE STATEMENT: We present a solution for creating merged public datasets supporting AI model development, using as an example the breast cancer domain and magnetic resonance imaging images. KEY POINTS: • The proposed platform allows unified access to the largest, homogenized public imaging dataset for breast cancer. • A methodology for the semantically enriched homogenization of public clinical data is presented. • The platform is able to make a detailed selection of breast MRI data for the development of AI models.

Prediction of immunochemotherapy response for diffuse large B-cell lymphoma using artificial intelligence digital pathology.

Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous and prevalent subtype of aggressive non-Hodgkin lymphoma that poses diagnostic and prognostic challenges, particularly in predicting drug responsiveness. In this study, we used digital pathology and deep learning to predict responses to immunochemotherapy in patients with DLBCL. We retrospectively collected 251 slide images from 216 DLBCL patients treated with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP), with their immunochemotherapy response labels. The digital pathology images were processed using contrastive learning for feature extraction. A multi-modal prediction model was developed by integrating clinical data and pathology image features. Knowledge distillation was employed to mitigate overfitting on gigapixel histopathology images to create a model that predicts responses based solely on pathology images. Based on the importance derived from the attention mechanism of the model, we extracted histological features that were considered key textures associated with drug responsiveness. The multi-modal prediction model achieved an impressive area under the ROC curve of 0.856, demonstrating significant associations with clinical variables such as Ann Arbor stage, International Prognostic Index, and bulky disease. Survival analyses indicated their effectiveness in predicting relapse-free survival. External validation using TCGA datasets supported the model's ability to predict survival differences. Additionally, pathology-based predictions show promise as independent prognostic indicators. Histopathological analysis identified centroblastic and immunoblastic features to be associated with treatment response, aligning with previous morphological classifications and highlighting the objectivity and reproducibility of artificial intelligence-based diagnosis. This study introduces a novel approach that combines digital pathology and clinical data to predict the response to immunochemotherapy in patients with DLBCL. This model shows great promise as a diagnostic and prognostic tool for clinical management of DLBCL. Further research and genomic data integration hold the potential to enhance its impact on clinical practice, ultimately improving patient outcomes.

Association Between Body Composition and Survival in Patients With Gastroesophageal Adenocarcinoma: An Automated Deep Learning Approach.

PURPOSE: Body composition (BC) may play a role in outcome prognostication in patients with gastroesophageal adenocarcinoma (GEAC). Artificial intelligence provides new possibilities to opportunistically quantify BC from computed tomography (CT) scans. We developed a deep learning (DL) model for fully automatic BC quantification on routine staging CTs and determined its prognostic role in a clinical cohort of patients with GEAC. MATERIALS AND METHODS: We developed and tested a DL model to quantify BC measures defined as subcutaneous and visceral adipose tissue (VAT) and skeletal muscle on routine CT and investigated their prognostic value in a cohort of patients with GEAC using baseline, 3-6-month, and 6-12-month postoperative CTs. Primary outcome was all-cause mortality, and secondary outcome was disease-free survival (DFS). Cox regression assessed the association between (1) BC at baseline and mortality and (2) the decrease in BC between baseline and follow-up scans and mortality/DFS. RESULTS: Model performance was high with Dice coefficients ≥0.94 ± 0.06. Among 299 patients with GEAC (age 63.0 ± 10.7 years; 19.4% female), 140 deaths (47%) occurred over a median follow-up of 31.3 months. At baseline, no BC measure was associated with DFS. Only a substantial decrease in VAT >70% after a 6- to 12-month follow-up was associated with mortality (hazard ratio [HR], 1.99 [95% CI, 1.18 to 3.34]; P = .009) and DFS (HR, 1.73 [95% CI, 1.01 to 2.95]; P = .045) independent of age, sex, BMI, Union for International Cancer Control stage, histologic grading, resection status, neoadjuvant therapy, and time between surgery and follow-up CT. CONCLUSION: DL enables opportunistic estimation of BC from routine staging CT to quantify prognostic information. In patients with GEAC, only a substantial decrease of VAT 6-12 months postsurgery was an independent predictor for DFS beyond traditional risk factors, which may help to identify individuals at high risk who go otherwise unnoticed.

Utility of artificial intelligence for detection of pneumothorax on chest radiopgraphs done after transthoracic percutaneous transthoracic biopsy guided by computed tomography.

OBJETIVE: To assess the ability of an artificial intelligence software to detect pneumothorax in chest radiographs done after percutaneous transthoracic biopsy. MATERIAL AND METHODS: We included retrospectively in our study adult patients who underwent CT-guided percutaneous transthoracic biopsies from lung, pleural or mediastinal lesions from June 2019 to June 2020, and who had a follow-up chest radiograph after the procedure. These chest radiographs were read to search the presence of pneumothorax independently by an expert thoracic radiologist and a radiodiagnosis resident, whose unified lecture was defined as the gold standard, and the result of each radiograph after interpretation by the artificial intelligence software was documented for posterior comparison with the gold standard. RESULTS: A total of 284 chest radiographs were included in the study and the incidence of pneumothorax was 14.4%. There were no discrepancies between the two readers' interpretation of any of the postbiopsy chest radiographs. The artificial intelligence software was able to detect 41/41 of the present pneumothorax, implying a sensitivity of 100% and a negative predictive value of 100%, with a specificity of 79.4% and a positive predictive value of 45%. The accuracy was 82.4%, indicating that there is a high probability that an individual will be adequately classified by the software. It has also been documented that the presence of Port-a-cath is the cause of 8 of the 50 of false positives by the software. CONCLUSIONS: The software has detected 100% of cases of pneumothorax in the postbiopsy chest radiographs. A potential use of this software could be as a prioritisation tool, allowing radiologists not to read immediately (or even not to read) chest radiographs classified as non-pathological by the software, with the confidence that there are no pathological cases.

Early screening of colorectal cancer using feature engineering with artificial intelligence-enhanced analysis of nanoscale chromatin modifications.

Colonoscopy is accurate but inefficient for colorectal cancer (CRC) prevention due to the low (~ 7 to 8%) prevalence of target lesions, advanced adenomas. We leveraged rectal mucosa to identify patients who harbor CRC field carcinogenesis by evaluating chromatin 3D architecture. Supranucleosomal disordered chromatin chains (~ 5 to 20 nm, ~1 kbp) fold into chromatin packing domains (~ 100 to 200 nm, ~ 100 to 1000 kbp). In turn, the fractal-like conformation of DNA within chromatin domains and the folding of the genome into packing domains has been shown to influence multiple facets of gene transcription, including the transcriptional plasticity of cancer cells. We deployed an optical spectroscopic nanosensing technique, chromatin-sensitive partial wave spectroscopic microscopy (csPWS), to evaluate the packing density scaling D of the chromatin chain conformation within packing domains from rectal mucosa in 256 patients with varying degrees of progression to colorectal cancer. We found average packing scaling D of chromatin domains was elevated in tumor cells, histologically normal-appearing cells 4 cm proximal to the tumor, and histologically normal-appearing rectal mucosa compared to cells from control patients (p < 0.001). Nuclear D had a robust correlation with the model of 5-year risk of CRC with r2 = 0.94. Furthermore, rectal D was evaluated as a screening biomarker for patients with advanced adenomas presenting an AUC of 0.85 and 85% sensitivity and specificity. artificial intelligence-enhanced csPWS improved diagnostic performance with AUC = 0.90. Considering the low sensitivity of existing CRC tests, including liquid biopsies, to early-stage cancers our work highlights the potential of chromatin biomarkers of field carcinogenesis in detecting early, significant precancerous colon lesions.

Multimodal modeling with low-dose CT and clinical information for diagnostic artificial intelligence on mediastinal tumors: a preliminary study.

BACKGROUND: Diagnosing mediastinal tumours, including incidental lesions, using low-dose CT (LDCT) performed for lung cancer screening, is challenging. It often requires additional invasive and costly tests for proper characterisation and surgical planning. This indicates the need for a more efficient and patient-centred approach, suggesting a gap in the existing diagnostic methods and the potential for artificial intelligence technologies to address this gap. This study aimed to create a multimodal hybrid transformer model using the Vision Transformer that leverages LDCT features and clinical data to improve surgical decision-making for patients with incidentally detected mediastinal tumours. METHODS: This retrospective study analysed patients with mediastinal tumours between 2010 and 2021. Patients eligible for surgery (n=30) were considered 'positive,' whereas those without tumour enlargement (n=32) were considered 'negative.' We developed a hybrid model combining a convolutional neural network with a transformer to integrate imaging and clinical data. The dataset was split in a 5:3:2 ratio for training, validation and testing. The model's efficacy was evaluated using a receiver operating characteristic (ROC) analysis across 25 iterations of random assignments and compared against conventional radiomics models and models excluding clinical data. RESULTS: The multimodal hybrid model demonstrated a mean area under the curve (AUC) of 0.90, significantly outperforming the non-clinical data model (AUC=0.86, p=0.04) and radiomics models (random forest AUC=0.81, p=0.008; logistic regression AUC=0.77, p=0.004). CONCLUSION: Integrating clinical and LDCT data using a hybrid transformer model can improve surgical decision-making for mediastinal tumours, showing superiority over models lacking clinical data integration.