Deep Learning Predicts Lung Cancer Treatment Response from Serial Medical Imaging

A multimodal deep learning model for detecting endoscopic images of near-infrared fluorescence capsules

Early screening for gastrointestinal (GI) diseases is critical for preventing cancer development. With the rapid advancement of deep learning technology, artificial intelligence (AI) has become increasingly prominent in the early detection of GI diseases. Capsule endoscopy is a non-invasive medical imaging technique used to examine the gastrointestinal tract. In our previous work, we developed a near-infrared fluorescence capsule endoscope (NIRF-CE) capable of exciting and capturing near-infrared (NIR) fluorescence images to specifically identify subtle mucosal microlesions and submucosal abnormalities while simultaneously capturing conventional white-light images to detect lesions with significant morphological changes. However, limitations such as low camera resolution and poor lighting within the gastrointestinal tract may lead to misdiagnosis and other medical errors. Manually reviewing and interpreting large volumes of capsule endoscopy images is time-consuming and prone to errors. Deep learning models have shown potential in automatically detecting abnormalities in NIRF-CE images. This study focuses on an improved deep learning model called Retinex-Attention-YOLO (RAY), which is based on single-modality image data and built on the YOLO series of object detection models. RAY enhances the accuracy and efficiency of anomaly detection, especially under low-light conditions. To further improve detection performance, we also propose a multimodal deep learning model, Multimodal-Retinex-Attention-YOLO (MRAY), which combines both white-light and fluorescence image data. The dataset used in this study consists of images of pig stomachs captured by our NIRF-CE system, simulating the human GI tract. In conjunction with a targeted fluorescent probe, which accumulates at lesion sites and releases fluorescent signals for imaging when abnormalities are present, a bright spot indicates a lesion. The MRAY model achieved an impressive precision of 96.3%, outperforming similar object detection models. To further validate the model's performance, ablation experiments were conducted, and comparisons were made with publicly available datasets. MRAY shows great promise for the automated detection of GI cancers, ulcers, inflammations, and other medical conditions in clinical practice.

Genetic variants of FER and SULF1 in the fibroblast-related genes are associated with non-small-cell lung cancer survival

Fibroblasts are important components in the tumor microenvironment and can affect tumor progression and metastasis. However, the roles of genetic variants of the fibroblast-related genes (FRGs) in the prognosis of non-small-cell lung cancer (NSCLC) patients have not been reported. Therefore, we investigated the associations between 26,544 single nucleotide polymorphisms (SNPs) in 291 FRGs and survival of NSCLC patients from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial. In Cox regression multivariable analysis, we found that 661 SNPs were associated with NSCLC overall survival (OS). Then we validated these SNPs in another independent replication dataset of 984 patients from the Harvard Lung Cancer Susceptibility (HLCS) Study. Finally, we identified two independent SNPs (i.e., FER rs7716388 A>G and SULF1 rs11785839 G>C) that remained significantly associated with NSCLC survival with hazards ratios (HRs) of 0.87 (95% confidence interval [CI] = 0.77-0.98, p = 0.018) and 0.88 (95% CI = 0.79-0.99, p = 0.033), respectively. Combined analysis for these two SNPs showed that the number of protective alleles was associated with better OS and disease-specific survival. Expression quantitative trait loci analysis indicated that the FER rs7716388 G allele was associated with the up-regulation of FER mRNA expression levels in lung tissue. Our results indicated that these two functional SNPs in the FRGs may be prognostic biomarkers for the prognosis of NSCLC patients, and the possible mechanism may be through modulating the expression of their corresponding genes.

Predicting postoperative prognosis in clear cell renal cell carcinoma using a multiphase CT-based deep learning model

BACKGROUND

Some clinicopathological risk stratification systems (CRSSs) such as the leibovich score have been used to predict the postoperative prognosis of patients with clear cell renal cell carcinoma (ccRCC), but there are no reliable noninvasive preoperative indicators for predicting postoperative prognosis in clinical practice.

PURPOSE

To assess the value of a deep learning (DL) model based on CT images in predicting the postoperative prognosis of patients with ccRCC.

MATERIALS AND METHODS

A total of 382 patients with ccRCC were retrospectively enrolled andallocated to training (n = 229) or testing (n = 153) cohorts at a 6:4 ratio. The features were extracted from precontrast-phase (PCP), corticomedullary-phase (CMP) and nephrographic-phase (NP) CT images with ResNet50, and then extreme learning machines (ELMs) were used to construct classification models. The DL model and Leibovich score were compared and combined. A receiver operating characteristic (ROC) curve and integrated discrimination improvement (IDI) were used to evaluate model performance.

RESULTS

Compared with other single-phase DL models, the three-phase CT-based DL model achieved the best performance, with an area under the curve (AUC) of 0.839. Combining the three-phase DL model and the Leibovich score (AUC = 0.823) into a nomogram (AUC = 0.888) statistically improved performance (IDINomogram vs. Three-phase = 0.1358, IDINomogram vs. Leibovich = 0.1393, [Formula: see text]< 0.001).

CONCLUSION

The CT-based DL model could be valuable for preoperatively predicting the prognosis of patients with ccRCC, and combining it with the Leibovich score can further improve its predictive performance.

Multicenter evaluation of predictive clinical and imaging factors for pathological response in non-small cell lung cancer patients treated with neoadjuvant chemotherapy and immune checkpoint inhibitors

BACKGROUND

This study aimed to identify clinical factors and develop a predictive model for pathological complete response (pCR) and major pathological response (MPR) in non-small cell lung cancer (NSCLC) patients receiving neoadjuvant chemotherapy combined with immune checkpoint inhibitors (ICIs).

METHODS

Cases meeting inclusion criteria were divided into high- and low-risk groups according to 75 clinical indicators based on tenfold LASSO selection. Logistic regression was employed to analyze both pCR and MPR. The accuracy of the nomograms was assessed using the time-dependent area under the curve (AUC).

RESULTS

A total of 297 patients from four multiple centers were included in the study, with 212 assigned to the training set and 85 to the testing set. The AUC was determined for the prediction of pCR (training: 0.97; testing: 0.88) and MPR (training: 0.98; testing: 0.81). Significant associations were observed between the preoperative tumor maximum diameter, preoperative tumor maximum standardized uptake value (SUVmax), changes in tumor SUVmax, percentage of tumor reduction, baseline total prostate-specific antigen (TPSA) and pathological response (P < 0.001).

CONCLUSIONS

The combined application of clinical indicators including non-invasive tumor imaging and hematology can help clinicians to obtain a higher ability to predict NSCLC patient's pathological remission, and the effect is better than that of clinical factors alone. These findings could help guide personalized treatment strategies in this patient population.

Empowering cancer prevention with AI: unlocking new frontiers in prediction, diagnosis, and intervention

Artificial intelligence is rapidly changing our world at an exponential rate and its transformative power has extensively reached important sectors like healthcare. In the fight against cancer, AI proved to be a novel and powerful tool, offering new hope for prevention and early detection. In this review, we will comprehensively explore the medical applications of AI, including early cancer detection through pathological and imaging analysis, risk stratification, patient triage, and the development of personalized prevention approaches. However, despite the successful impact AI has contributed to, we will also discuss the myriad of challenges that we have faced so far toward optimal AI implementation. There are problems when it comes to the best way in which we can use AI systemically. Having the correct data that can be understood easily must remain one of the most significant concerns in all its uses including sharing information. Another challenge that exists is how to interpret AI models because they are too complicated for people to follow through examples used in their developments which may affect trust, especially among medical professionals. Other considerations like data privacy, algorithm bias, and equitable access to AI tools have also arisen. Finally, we will evaluate possible future directions for this promising field that highlight AI's capacity to transform preventative cancer care.

Current and Future Applications of PET Radiomics in Radiation Oncology

This review delves into the principles of PET imaging and radiomics, emphasizing their importance in detecting, staging, and monitoring various cancers. It highlights the clinical applications of PET radiomics in oncology, showcasing its impact on personalized cancer care. Additionally, the review addresses challenges such as standardizing PET radiomics, integrating multiomics data, and ethical concerns in clinical decision-making. Future directions are also discussed, including broader applications of PET radiomics in clinical trials, artificial intelligence integration for automated analysis, and incorporating multiomics data for a comprehensive understanding of tumor biology.

Critical review of patient outcome study in head and neck cancer radiotherapy

Rapid technological advances in radiation therapy have significantly improved dose delivery and tumor control for head and neck cancers. However, treatment-related toxicities caused by high-dose exposure to critical structures remain a significant clinical challenge, underscoring the need for accurate prediction of clinical outcomes-encompassing both tumor control and adverse events (AEs). This review critically evaluates the evolution of data-driven approaches in predicting patient outcomes in head and neck cancer patients treated with radiation therapy, from traditional dose-volume constraints to cutting-edge artificial intelligence (AI) and causal inference framework. The integration of linear energy transfer in patient outcomes study, which has uncovered critical mechanisms behind unexpected toxicity, was also introduced for proton therapy. Three transformative methodological advances are reviewed: radiomics, AI-based algorithms, and causal inference frameworks. While radiomics has enabled quantitative characterization of medical images, AI models have demonstrated superior capability than traditional models. However, the field faces significant challenges in translating statistical correlations from real-world data into interventional clinical insights. We highlight that how causal inference methods can bridge this gap by providing a rigorous framework for identifying treatment effects. Looking ahead, we envision that combining these complementary approaches, especially the interventional prediction models, will enable more personalized treatment strategies, ultimately improving both tumor control and quality of life for head and neck cancer patients treated with radiation therapy.

Establishing a Deep Learning Model That Integrates Pretreatment and Midtreatment Computed Tomography to Predict Treatment Response in Non-Small Cell Lung Cancer

PURPOSE

Patients with identical stages or similar tumor volumes can vary significantly in their responses to radiation therapy (RT) due to individual characteristics, making personalized RT for non-small cell lung cancer (NSCLC) challenging. This study aimed to develop a deep learning model by integrating pretreatment and midtreatment computed tomography (CT) to predict the treatment response in NSCLC patients.

METHODS AND MATERIALS

We retrospectively collected data from 168 NSCLC patients across 3 hospitals. Data from Shanghai General Hospital (SGH, 35 patients) and Shanxi Cancer Hospital (SCH, 93 patients) were used for model training and internal validation, while data from Linfen Central Hospital (LCH, 40 patients) were used for external validation. Deep learning, radiomics, and clinical features were extracted to establish a varying time interval long short-term memory network for response prediction. Furthermore, we derived a model-deduced personalize dose escalation (DE) for patients predicted to have suboptimal gross tumor volume regression. The area under the receiver operating characteristic curve (AUC) and predicted absolute error were used to evaluate the predictive Response Evaluation Criteria in Solid Tumors classification and the proportion of gross tumor volume residual. DE was calculated as the biological equivalent dose using an /α/β ratio of 10 Gy.

RESULTS

The model using only pretreatment CT achieved the highest AUC of 0.762 and 0.687 in internal and external validation respectively, whereas the model integrating both pretreatment and midtreatment CT achieved AUC of 0.869 and 0.798, with predicted absolute error of 0.137 and 0.185, respectively. We performed personalized DE for 29 patients. Their original biological equivalent dose was approximately 72 Gy, within the range of 71.6 Gy to 75 Gy. DE ranged from 77.7 to 120 Gy for 29 patients, with 17 patients exceeding 100 Gy and 8 patients reaching the model's preset upper limit of 120 Gy.

CONCLUSIONS

Combining pretreatment and midtreatment CT enhances prediction performance for RT response and offers a promising approach for personalized DE in NSCLC.

The 100 most-cited radiomics articles in cancer research: A bibliometric analysis

Radiomics, an advanced medical imaging analysis technique introduced by Professor Lambin in 2012, has quickly become a key area of medical research. To explore emerging trends in cancer-related radiomics, we conducted a bibliometric analysis of the 100 most-cited articles (T100) in this field. Data were collected from the Web of Science Core Collection on October 7, 2023, and the articles were ranked by citation count. We extracted data such as authors, journals, citation counts, and publication years and analyzed it using Microsoft Excel 2019 and R 4.4.2. CiteSpace was used to create co-occurrence and citation burst maps to show the relationships between authors, countries, institutions, and keywords. The analysis revealed that the T100 came from 81 countries, with the U.S. contributing the most (56 articles). Harvard University was the leading institution, and the journal Radiology had the highest citation count. Aerts Hugo JWL was the most influential author. The study highlights that "lung cancer" and "artificial intelligence" are emerging as major research hotspots, shaping the future of cancer radiomics.

The role of radiomics for predicting of lymph-vascular space invasion in cervical cancer patients based on artificial intelligence: a systematic review and meta-analysis

The primary aim of this study was to conduct a methodical examination and assessment of the prognostic efficacy exhibited by magnetic resonance imaging (MRI)-derived radiomic models concerning the preoperative prediction of lymph-vascular space infiltration (LVSI) in cervical cancer cases. A comprehensive and thorough exploration of pertinent academic literature was undertaken by two investigators, employing the resources of the Embase, PubMed, Web of Science, and Cochrane Library databases. The scope of this research was bounded by a publication cutoff date of May 15, 2023. The inclusion criteria encompassed studies that utilized radiomic models based on MRI to prognosticate the accuracy of preoperative LVSI estimation in instances of cervical cancer. The Diagnostic Accuracy Studies-2 framework and the Radiomic Quality Score metric were employed. This investigation included nine distinct research studies, enrolling a total of 1,406 patients. The diagnostic performance metrics of MRI-based radiomic models in the prediction of preoperative LVSI among cervical cancer patients were determined as follows: sensitivity of 83% (95% confidence interval [CI]=77%-87%), specificity of 74% (95% CI=69%-79%), and a corresponding AUC of summary receiver operating characteristic measuring 0.86 (95% CI=0.82-0.88). The results of the synthesized meta-analysis did not reveal substantial heterogeneity.This meta-analysis suggests the robust diagnostic proficiency of the MRI-based radiomic model in the prognostication of preoperative LVSI within the cohort of cervical cancer patients. In the future, radiomics holds the potential to emerge as a widely applicable noninvasive modality for the early detection of LVSI in the context of cervical cancer.

Towards the Prediction of Responses to Cancer Immunotherapy: A Multi-Omics Review

Tumor treatment has undergone revolutionary changes with the development of immunotherapy, especially immune checkpoint inhibitors. Because not all patients respond positively to immune therapeutic agents, and severe immune-related adverse events (irAEs) are frequently observed, the development of the biomarkers evaluating the response of a patient is key for the application of immunotherapy in a wider range. Recently, various multi-omics features measured by high-throughput technologies, such as tumor mutation burden (TMB), gene expression profiles, and DNA methylation profiles, have been proved to be sensitive and accurate predictors of the response to immunotherapy. A large number of predictive models based on these features, utilizing traditional machine learning or deep learning frameworks, have also been proposed. In this review, we aim to cover recent advances in predicting tumor immunotherapy response using multi-omics features. These include new measurements, research cohorts, data sources, and predictive models. Key findings emphasize the importance of TMB, neoantigens, MSI, and mutational signatures in predicting ICI responses. The integration of bulk and single-cell RNA sequencing has enhanced our understanding of the tumor immune microenvironment and enabled the identification of predictive biomarkers like PD-L1 and IFN-γ signatures. Public datasets and machine learning models have also improved predictive tools. However, challenges remain, such as the need for large and diverse clinical datasets, standardization of multi-omics data, and model interpretability. Future research will require collaboration among researchers, clinicians, and data scientists to address these issues and enhance cancer immunotherapy precision.

Exploring the role of artificial intelligence in chemotherapy development, cancer diagnosis, and treatment: present achievements and future outlook

Background

Artificial intelligence (AI) has emerged as a transformative tool in oncology, offering promising applications in chemotherapy development, cancer diagnosis, and predicting chemotherapy response. Despite its potential, debates persist regarding the predictive accuracy of AI technologies, particularly machine learning (ML) and deep learning (DL).

Objective

This review aims to explore the role of AI in forecasting outcomes related to chemotherapy development, cancer diagnosis, and treatment response, synthesizing current advancements and identifying critical gaps in the field.

Methods

A comprehensive literature search was conducted across PubMed, Embase, Web of Science, and Cochrane databases up to 2023. Keywords included "Artificial Intelligence (AI)," "Machine Learning (ML)," and "Deep Learning (DL)" combined with "chemotherapy development," "cancer diagnosis," and "cancer treatment." Articles published within the last four years and written in English were included. The Prediction Model Risk of Bias Assessment tool was utilized to assess the risk of bias in the selected studies.

Conclusion

This review underscores the substantial impact of AI, including ML and DL, on cancer diagnosis, chemotherapy innovation, and treatment response for both solid and hematological tumors. Evidence from recent studies highlights AI's potential to reduce cancer-related mortality by optimizing diagnostic accuracy, personalizing treatment plans, and improving therapeutic outcomes. Future research should focus on addressing challenges in clinical implementation, ethical considerations, and scalability to enhance AI's integration into oncology care.

Assessing the Capabilities of Generative Pretrained Transformer-4 in Addressing Open-Ended Inquiries of Oral Cancer

INTRODUCTION AND AIMS

In the face of escalating oral cancer rates, the application of large language models like Generative Pretrained Transformer (GPT)-4 presents a novel pathway for enhancing public awareness about prevention and early detection. This research aims to explore the capabilities and possibilities of GPT-4 in addressing open-ended inquiries in the field of oral cancer.

METHODS

Using 60 questions accompanied by reference answers, covering concepts, causes, treatments, nutrition, and other aspects of oral cancer, evaluators from diverse backgrounds were selected to evaluate the capabilities of GPT-4 and a customized version. A P value under .05 was considered significant.

RESULTS

Analysis revealed that GPT-4 and its adaptations notably excelled in answering open-ended questions, with the majority of responses receiving high scores. Although the median score for standard GPT-4 was marginally better, statistical tests showed no significant difference in capabilities between the two models (P > .05). Despite statistical significance indicated diverse backgrounds of evaluators have statistically difference (P < .05), a post hoc test and comprehensive analysis demonstrated that both editions of GPT-4 demonstrated equivalent capabilities in answering questions concerning oral cancer.

CONCLUSIONS

GPT-4 has demonstrated its capability to furnish responses to open-ended inquiries concerning oral cancer. Utilizing this advanced technology to boost public awareness about oral cancer is viable and has much potential. When it's unable to locate pertinent information, it will resort to their inherent knowledge base or recommend consulting professionals after offering some basic information. Therefore, it cannot supplant the expertise and clinical judgment of surgical oncologists and could be used as an adjunctive evaluation tool.

A Serial MRI-based Deep Learning Model to Predict Survival in Patients with Locoregionally Advanced Nasopharyngeal Carcinoma

Purpose To develop and evaluate a deep learning-based prognostic model for predicting survival in locoregionally advanced nasopharyngeal carcinoma (LA-NPC) using serial MRI before and after induction chemotherapy (IC). Materials and Methods This multicenter retrospective study included 1039 patients with LA-NPC (779 male and 260 female patients; mean age, 44 years ± 11 [SD]) diagnosed between December 2011 and January 2016. A radiomics-clinical prognostic model (model RC) was developed using pre- and post-IC MRI acquisitions and other clinical factors using graph convolutional neural networks. The concordance index (C-index) was used to evaluate model performance in predicting disease-free survival (DFS). The survival benefits of concurrent chemoradiation therapy (CCRT) were analyzed in model-defined risk groups. Results The C-indexes of model RC for predicting DFS were significantly higher than those of TNM staging in the internal (0.79 vs 0.53) and external (0.79 vs 0.62, both P < .001) testing cohorts. The 5-year DFS for the model RC-defined low-risk group was significantly better than that of the high-risk group (90.6% vs 58.9%, P < .001). In high-risk patients, those who underwent CCRT had a higher 5-year DFS rate than those who did not (58.7% vs 28.6%, P = .03). There was no evidence of a difference in 5-year DFS rate in low-risk patients who did or did not undergo CCRT (91.9% vs 81.3%, P = .19). Conclusion Serial MRI before and after IC can effectively help predict survival in LA-NPC. The radiomics-clinical prognostic model developed using a graph convolutional network-based deep learning method showed good risk discrimination capabilities and may facilitate risk-adapted therapy. Keywords: Nasopharyngeal Carcinoma, Deep Learning, Induction Chemotherapy, Serial MRI, MR Imaging, Radiomics, Prognosis, Radiation Therapy/Oncology, Head/Neck Supplemental material is available for this article. © RSNA, 2025.

Longitudinal Image Data for Outcome Modeling

In oncology, medical imaging is crucial for diagnosis, treatment planning and therapy execution. Treatment responses can be complex and varied and are known to involve factors of treatment, patient characteristics and tumor microenvironment. Longitudinal image analysis is able to track temporal changes, aiding in disease monitoring, treatment evaluation, and outcome prediction. This allows for the enhancement of personalized medicine. However, analyzing longitudinal 2D and 3D images presents unique challenges, including image registration, reliable segmentation, dealing with variable imaging intervals, and sparse data. This review presents an overview of techniques and methodologies in longitudinal image analysis, with a primary focus on outcome modeling in radiation oncology.

A Holistic Approach to Implementing Artificial Intelligence in Lung Cancer

The application of artificial intelligence (AI) in lung cancer, particularly in surgical approaches, has significantly transformed the healthcare landscape. AI has demonstrated remarkable advancements in early lung cancer detection, precise medical image analysis, and personalized treatment planning, all of which are crucial for surgical interventions. By analyzing extensive datasets, AI algorithms can identify patterns and anomalies in lung scans, facilitating timely diagnoses and enhancing surgical outcomes. Furthermore, AI can detect subtle indicators that may be overlooked by human practitioners, leading to quicker intervention and more effective treatment strategies. The technology can also predict patient responses to surgical treatments, enabling tailored care plans that improve recovery rates. In addition to surgical applications, AI streamlines administrative tasks such as record management and appointment scheduling, allowing healthcare providers to concentrate on delivering high-quality care. The integration of AI with genomics and precision medicine holds the potential to further refine surgical approaches in lung cancer treatment by developing targeted strategies that enhance effectiveness and minimize side effects. Despite challenges related to data privacy and regulatory concerns, the ongoing advancements in AI, coupled with collaboration between healthcare professionals and AI experts, suggest a promising future for lung cancer care. This article explores how AI addresses the challenges of lung cancer treatment, focusing on current advancements, obstacles, and the future potential of surgical applications.

Lung cancer diagnosis with GAN supported deep learning models

BACKGROUND

Lung cancer is a leading cause of cancer-related deaths worldwide, making early diagnosis crucial for improving treatment success and survival rates. Traditional diagnostic methods, such as biopsy and manual CT image interpretation, are time-consuming and prone to variability, highlighting the need for more efficient and accurate tools. Advances in deep learning offer promising solutions by enabling faster and more objective medical image analysis.

OBJECTIVE

This study aims to classify benign, malignant, and normal lung CT images using advanced deep learning techniques, including a specially developed CNN model, to improve diagnostic accuracy.

METHODS

A dataset of 1097 lung CT images was balanced using GANs and preprocessed with techniques like histogram equalization and noise reduction. The data was split into 70% training and 30% testing sets. Models including VGG19, AlexNet, InceptionV3, ResNet50, and a custom-designed CNN were trained. Additionally, Faster R-CNN-based region proposal methods were integrated to enhance detection performance.

RESULTS

The custom CNN model achieved the highest accuracy at 99%, surpassing other architectures like VGG19, which reached 97%. The Faster R-CNN integration further improved sensitivity and classification precision.

CONCLUSION

The results demonstrate the effectiveness of GAN-supported deep learning models for lung cancer classification, highlighting their potential clinical applications for early detection and diagnosis.

Exploring machine learning tools in a retrospective case-study of patients with metastatic non-small cell lung cancer treated with first-line immunotherapy: A feasibility single-centre experience

BACKGROUND

Artificial intelligence (AI) models are emerging as promising tools to identify predictive features among data coming from health records. Their application in clinical routine is still challenging, due to technical limits and to explainability issues in this specific setting. Response to standard first-line immunotherapy (ICI) in metastatic Non-Small-Cell Lung Cancer (NSCLC) is an interesting population for machine learning (ML), since up to 30% of patients do not benefit.

METHODS

We retrospectively collected all consecutive patients with PD-L1 ≥ 50 % metastatic NSCLC treated with first-line ICI at our institution between 2017 and 2021. Demographic, laboratory, molecular and clinical data were retrieved manually or automatically according to data sources. Primary aim was to explore feasibility of ML models in clinical routine setting and to detect problems and solutions for everyday implementation. Early progression was used as preliminary endpoint to test our algorithm.

RESULTS

Out of 123 patients, 106 were included, 52/106 (49 %) had disease progression or died within 3 months of start of ICI. Early progression correlated with increased neutrophil percentage (>80 % of white blood cells), neutrophil/lymphocyte ratio (≥8) and lower-range PD-L1 status (<70 %) at baseline, which was consistent with literature. Automated ML (AutoML) models run on our dataset reached precision scores around 80 %, with Voting Ensemble emerging as best performing model, while white-box models (such as Shapley Additive exPlanations) provided better explainability. In all AutoML models, laboratory features were the top selected features, whilst clinical ones needed more pre-processing before gaining relevance, which was consistent with different data extraction (automatic versus manual) and missing data rates.

CONCLUSIONS

ML models' application is feasible in clinical practice and can trustworthily predict early progression during first-line ICI for metastatic NSCLC. Solving pre-analytical issues is key for future improvement, focusing on automatic tools for data extraction, collection and explainability.

Deep learning in pulmonary nodule detection and segmentation: a systematic review

OBJECTIVES

The accurate detection and precise segmentation of lung nodules on computed tomography are key prerequisites for early diagnosis and appropriate treatment of lung cancer. This study was designed to compare detection and segmentation methods for pulmonary nodules using deep-learning techniques to fill methodological gaps and biases in the existing literature.

METHODS

This study utilized a systematic review with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines, searching PubMed, Embase, Web of Science Core Collection, and the Cochrane Library databases up to May 10, 2023. The Quality Assessment of Diagnostic Accuracy Studies 2 criteria was used to assess the risk of bias and was adjusted with the Checklist for Artificial Intelligence in Medical Imaging. The study analyzed and extracted model performance, data sources, and task-focus information.

RESULTS

After screening, we included nine studies meeting our inclusion criteria. These studies were published between 2019 and 2023 and predominantly used public datasets, with the Lung Image Database Consortium Image Collection and Image Database Resource Initiative and Lung Nodule Analysis 2016 being the most common. The studies focused on detection, segmentation, and other tasks, primarily utilizing Convolutional Neural Networks for model development. Performance evaluation covered multiple metrics, including sensitivity and the Dice coefficient.

CONCLUSIONS

This study highlights the potential power of deep learning in lung nodule detection and segmentation. It underscores the importance of standardized data processing, code and data sharing, the value of external test datasets, and the need to balance model complexity and efficiency in future research.

CLINICAL RELEVANCE STATEMENT

Deep learning demonstrates significant promise in autonomously detecting and segmenting pulmonary nodules. Future research should address methodological shortcomings and variability to enhance its clinical utility.

KEY POINTS

Deep learning shows potential in the detection and segmentation of pulmonary nodules. There are methodological gaps and biases present in the existing literature. Factors such as external validation and transparency affect the clinical application.

Descriptive overview of AI applications in x-ray imaging and radiotherapy

Artificial intelligence (AI) is transforming medical radiation applications by handling complex data, learning patterns, and making accurate predictions, leading to improved patient outcomes. This article examines the use of AI in optimising radiation doses for x-ray imaging, improving radiotherapy outcomes, and briefly addresses the benefits, challenges, and limitations of AI integration into clinical workflows. In diagnostic radiology, AI plays a pivotal role in optimising radiation exposure, reducing noise, enhancing image contrast, and lowering radiation doses, especially in high-dose procedures like computed tomography (CT). Deep learning (DL)-powered CT reconstruction methods have already been incorporated into clinical routine. Moreover, AI-powered methodologies have been developed to provide real-time, patient-specific radiation dose estimates. These AI-driven tools have the potential to streamline workflows and potentially become integral parts of imaging practices. In radiotherapy, AI's ability to automate and enhance the precision of treatment planning is emphasised. Traditional methods, such as manual contouring, are time-consuming and prone to variability. AI-driven techniques, particularly DL models, are automating the segmentation of organs and tumours, improving the accuracy of radiation delivery, and minimising damage to healthy tissues. Moreover, AI supports adaptive radiotherapy, allowing continuous optimisation of treatment plans based on changes in a patient's anatomy over time, ensuring the highest accuracy in radiation delivery and better therapeutic outcomes. Some of these methods have been validated and integrated into radiation treatment systems, while others are not yet ready for routine clinical use mainly due to challenges in validation, particularly ensuring reliability across diverse patient populations and clinical settings. Despite the potential of AI, there are challenges in fully integrating these technologies into clinical practice. Issues such as data protection, privacy, data quality, model validation, and the need for large and diverse datasets are crucial to ensuring the reliability of AI systems.

Advancements in lung cancer: a comprehensive perspective on diagnosis, staging, therapy and follow-up from the SAKK Working Group on Imaging in Diagnosis and Therapy Monitoring

In 2015, around 4400 individuals received a diagnosis of lung cancer, and Switzerland recorded approximately 3200 deaths related to lung cancer. Advances in detection, such as lung cancer screening and improved treatments, have led to increased identification of early-stage lung cancer and higher chances of long-term survival. This progress has introduced new considerations in imaging, emphasising non-invasive diagnosis and characterisation techniques like radiomics. Treatment aspects, such as preoperative assessment and the implementation of immune response evaluation criteria in solid tumours (iRECIST), have also seen advancements. For those undergoing curative treatment for lung cancer, guidelines propose follow-up with computed tomography (CT) scans within a specific timeframe. However, discrepancies exist in published guidelines, and there is a lack of universally accepted recommendations for follow-up procedures. This white paper aims to provide a certain standard regarding the use of imaging on the diagnosis, staging, treatment and follow-up of patients with lung cancer.

The untapped potential of 3D virtualization using high resolution scanner-based and photogrammetry technologies for bone bank digital modeling

Three-dimensional (3D) scanning technologies could transform medical practices by creating virtual tissue banks. In bone transplantation, new approaches are needed to provide surgeons with accurate tissue measurements while minimizing contamination risks and avoiding repeated freeze-thaw cycles of banked tissues. This study evaluates three prominent non-contact 3D scanning methods-structured light scanning (SLG), laser scanning (LAS), and photogrammetry (PHG)-to support tissue banking operations. We conducted a thorough examination of each technology and the precision of the 3D scanned bones using relevant anatomical specimens under sterile conditions. Cranial caps were scanned as separate inner and outer surfaces, automatically aligned, and merged with post-processing. A colorimetric analysis based on CIEDE2000 was performed, and the results were compared with questionnaires distributed among neurosurgeons. The findings indicate that certain 3D scanning methods were more appropriate for specific bones. Among the technologies, SLG emerged as optimal for tissue banking, offering a superior balance of accuracy, minimal distortion, cost-efficiency, and ease of use. All methods slightly underestimated the volume of the specimens in their virtual models. According to the colorimetric analysis and the questionnaires given to the neurosurgeons, our low-cost PHG system performed better than others in capturing cranial caps, although it exhibited the least dimensional accuracy. In conclusion, this study provides valuable insights for surgeons and tissue bank personnel in selecting the most efficient 3D non-contact scanning technology and optimizing protocols for modernized tissue banking. Future work will advance towards smart healthcare solutions, explore the development of virtual tissue banks.

Deep learning of pretreatment multiphase CT images for predicting response to lenvatinib and immune checkpoint inhibitors in unresectable hepatocellular carcinoma

Objectives

Combination therapy of lenvatinib and immune checkpoint inhibitors (CLICI) has emerged as a promising approach for managing unresectable hepatocellular carcinoma (HCC). However, the response to such treatment is observed in only a subset of patients, underscoring the pressing need for reliable methods to identify potential responders.

Materials & methods

This was a retrospective analysis involving 120 patients with unresectable HCC. They were divided into training (n = 72) and validation (n = 48) cohorts. We developed an interpretable deep learning model using multiphase computed tomography (CT) images to predict whether patients will respond or not to CLICI treatment, based on the Response Evaluation Criteria in Solid Tumors, version 1.1 (RECIST v1.1). We evaluated the models' performance and analyzed the impact of each CT phase. Critical regions influencing predictions were identified and visualized through heatmaps.

Results

The multiphase model outperformed the best biphase and uniphase models, achieving an area under the curve (AUC) of 0.802 (95% CI = 0.780-0.824). The portal phase images were found to significantly enhance the model's predictive accuracy. Heatmaps identified six critical features influencing treatment response, offering valuable insights to clinicians. Additionally, we have made this model accessible via a web server at http://uhccnet.com/ for ease of use.

Conclusions

The integration of multiphase CT images with deep learning-generated heatmaps for predicting treatment response provides a robust and practical tool for guiding CLICI therapy in patients with unresectable HCC.

Predictive value of metabolic parameters and apparent diffusion coefficient derived from 18F-FDG PET/MR in patients with non-small cell lung cancer

BACKGROUND

Multiple models intravoxel incoherent motion (IVIM) based 18F-fluorodeoxyglucose positron emission tomography-magnetic resonance(18F-FDG PET/MR) could reflect the microscopic information of the tumor from multiple perspectives. However, its value in the prognostic assessment of non-small cell lung cancer (NSCLC) still needs to be further explored.

OBJECTIVE

To compare the value of 18F-FDG PET/MR metabolic parameters and diffusion parameters in the prognostic assessment of patients with NSCLC.

METERIAL AND METHODS

Chest PET and IVIM scans were performed on 61 NSCLC patients using PET/MR. The maximum standard uptake value (SUVmax), metabolic tumor volume (MTV), total lesion glycolysis (TLG), diffusion coefficient (D), perfusion fraction (f), pseudo diffusion coefficient (D*) and apparent diffusion coefficient (ADC) were calculated. The impact of SUVmax, MTV, TLG, D, f, D*and ADC on survival was measured in terms of the hazard ratio (HR) effect size. Overall survival time (OS) and progression-free survival time (PFS) were evaluated with the Kaplan-Meier and Cox proportional hazard models. Log-rank test was used to analyze the differences in parameters between groups.

RESULTS

61 NSCLC patients had an overall median OS of 18 months (14.75, 22.85) and a median PFS of 17 months (12.00, 21.75). Univariate analysis showed that pathological subtype, TNM stage, surgery, SUVmax, MTV, TLG, D, D* and ADC were both influential factors for OS and PFS in NSCLC patients. Multifactorial analysis showed that MTV, D* and ADC were independent predicting factors for OS and PFS in NSCLC patients.

CONCLUSION

MTV, D* and ADC are independent predicting factors affecting OS and PFS in NSCLC patients. 18F-FDG PET/MR-derived metabolic parameters and diffusion parameters have clinical value for prognostic assessment of NSCLC patients.

The prognostic value of radiogenomics using CT in patients with lung cancer: a systematic review

This systematic review aimed to evaluate the effectiveness of combining radiomic and genomic models in predicting the long-term prognosis of patients with lung cancer and to contribute to the further exploration of radiomics. This study retrieved comprehensive literature from multiple databases, including radiomics and genomics, to study the prognosis of lung cancer. The model construction consisted of the radiomic and genomic methods. A comprehensive bias assessment was conducted, including risk assessment and model performance indicators. Ten studies between 2016 and 2023 were analyzed. Studies were mostly retrospective. Patient cohorts varied in size and characteristics, with the number of patients ranging from 79 to 315. The construction of the model involves various radiomic and genotic datasets, and most models show promising prediction performance with the area under the receiver operating characteristic curve (AUC) values ranging from 0.64 to 0.94 and the concordance index (C-index) values from 0.28 to 0.80. The combination model typically outperforms the single method model, indicating higher prediction accuracy and the highest AUC was 0.99. Combining radiomics and genomics in the prognostic model of lung cancer may improve the predictive performance. However, further research on standardized data and larger cohorts is needed to validate and integrate these findings into clinical practice. CRITICAL RELEVANCE STATEMENT: The combination of radiomics and genomics in the prognostic model of lung cancer improved prediction accuracy in most included studies. KEY POINTS: The combination of radiomics and genomics can improve model performance in most studies. The results of establishing prognosis models by different methods are discussed. The combination of radiomics and genomics may be helpful to provide better treatment for patients.

Predicting drug outcome of population via clinical knowledge graph

Optimal treatments depend on numerous factors such as drug chemical properties, disease biology, and patient characteristics to which the treatment is applied. To realize the promise of AI in healthcare, there is a need for designing systems that can capture patient heterogeneity and relevant biomedical knowledge. Here we present PlaNet, a geometric deep learning framework that reasons over population variability, disease biology, and drug chemistry by representing knowledge in the form of a massive clinical knowledge graph that can be enhanced by language models. Our framework is applicable to any sub-population, any drug as well drug combinations, any disease, and a wide range of pharmacological tasks. We apply the PlaNet framework to reason about outcomes of clinical trials: PlaNet predicts drug efficacy and adverse events, even for experimental drugs and their combinations that have never been seen by the model. Furthermore, PlaNet can estimate the effect of changing population on trial outcomes with direct implications for patient stratification in clinical trials. PlaNet takes fundamental steps towards AI-guided clinical trials design, offering valuable guidance for realizing the vision of precision medicine using AI.

Advancements and Challenges in the Image-Based Diagnosis of Lung and Colon Cancer: A Comprehensive Review

Image-based diagnosis has become a crucial tool in the identification and management of various cancers, particularly lung and colon cancer. This review delves into the latest advancements and ongoing challenges in the field, with a focus on deep learning, machine learning, and image processing techniques applied to X-rays, CT scans, and histopathological images. Significant progress has been made in imaging technologies like computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography (PET), which, when combined with machine learning and artificial intelligence (AI) methodologies, have greatly enhanced the accuracy of cancer detection and characterization. These advances have enabled early detection, more precise tumor localization, personalized treatment plans, and overall improved patient outcomes. However, despite these improvements, challenges persist. Variability in image interpretation, the lack of standardized diagnostic protocols, unequal access to advanced imaging technologies, and concerns over data privacy and security within AI-based systems remain major obstacles. Furthermore, integrating imaging data with broader clinical information is crucial to achieving a more comprehensive approach to cancer diagnosis and treatment. This review provides valuable insights into the recent developments and challenges in image-based diagnosis for lung and colon cancers, underscoring both the remarkable progress and the hurdles that still need to be overcome to optimize cancer care.

Exploring clinical factors to predict the survival of patients with resectable non-small cell lung cancer with neoadjuvant immunotherapy

OBJECTIVES

The goal was to explore clinical factors and build a predictive model for the disease-free and overall survival of patients with non-small cell lung cancer (NSCLC) receiving neoadjuvant chemotherapy combined with immune checkpoint inhibitors.

METHODS

Inclusion criteria for patients in this multicentre study were as follows: (i) Patients who were diagnosed with stages I-III NSCLC after a bronchoscopy biopsy or puncture; (ii) patients who were examined with computed tomography/positron emission tomography-computed tomography before treatment and surgery; (iii) patients who received neoadjuvant chemotherapy combined with immune checkpoint inhibitors for 2 to 6 cycles preoperatively; (iv) patients whose peripheral blood indicators and tumour markers were assessed before treatment and preoperatively; (v) patients who underwent radical lung cancer surgery after neoadjuvant therapy. Cases were divided into high- and low-risk groups according to 78 clinical indicators based on a 10-fold Least Absolute Shrinkage and Selection Operator selection. We used Cox proportional hazards models to predict disease-free and overall survival. Then, we used time-dependent area under the curve and decision curve analyses to examine the accuracy of the results.

RESULTS

Data were collected continuously, and 212 and 85 cases were randomly assigned to training and testing sets, respectively. The area under the curve for the prediction of disease-free survival (training: 1 year, 0.83; 2 years, 0.81; 3 years, 0.83 versus testing: 1 year, 0.65; 2 years, 0.66; 3 years, 0.70), overall survival (training: 1 year, 0.86; 2 years, 0.85; 3 years, 0.86 versus testing: 1 year, 0.66; 2 years, 0.57; 3 years, 0.70) were determined. The coefficient factors including pathological response; preoperative tumour maximum diameter; preoperative lymph shorter diameter; preoperative tumour and lymph maximum standardized uptake value; change in tumour standardized uptake value preoperatively; and blood-related risk factors were favourably associated with prognosis (P < 0.001).

CONCLUSIONS

Our prediction model, which integrated data from preoperative positron emission tomography-CT, preoperative blood parameters and pathological response, was able to make highly accurate predictions for disease-free and overall survival in patients with NSCLC receiving neoadjuvant immunity with chemical therapy.

Advances in imaging modalities for spinal tumors

The spinal cord occupies a narrow region and is tightly surrounded by osseous and ligamentous structures; spinal tumors can damage this structure and deprive patients of their ability to independently perform activities of daily living. Hence, imaging is vital for the prompt detection and accurate diagnosis of spinal tumors, as well as determining the optimal treatment and follow-up plan. However, many clinicians may not be familiar with the imaging characteristics of spinal tumors due to their rarity. In addition, spinal surgeons might not fully utilize imaging for the surgical planning and management of spinal tumors because of the complex heterogeneity of these lesions. In the present review, we focus on conventional and advanced spinal tumor imaging techniques. These imaging modalities include computed tomography, positron emission tomography, digital subtraction angiography, conventional and microstructural magnetic resonance imaging, and high-resolution ultrasound. We discuss the advantages and disadvantages of conventional and emerging imaging modalities, followed by an examination of cutting-edge medical technology to complement current needs in the field of spinal tumors. Moreover, machine learning and artificial intelligence are anticipated to impact the application of spinal imaging techniques. Through this review, we discuss the importance of conventional and advanced spinal tumor imaging, and the opportunity to combine advanced technologies with conventional modalities to better manage patients with these lesions.

Artificial intelligence methods available for cancer research

Cancer is a heterogeneous and multifaceted disease with a significant global footprint. Despite substantial technological advancements for battling cancer, early diagnosis and selection of effective treatment remains a challenge. With the convenience of large-scale datasets including multiple levels of data, new bioinformatic tools are needed to transform this wealth of information into clinically useful decision-support tools. In this field, artificial intelligence (AI) technologies with their highly diverse applications are rapidly gaining ground. Machine learning methods, such as Bayesian networks, support vector machines, decision trees, random forests, gradient boosting, and K-nearest neighbors, including neural network models like deep learning, have proven valuable in predictive, prognostic, and diagnostic studies. Researchers have recently employed large language models to tackle new dimensions of problems. However, leveraging the opportunity to utilize AI in clinical settings will require surpassing significant obstacles-a major issue is the lack of use of the available reporting guidelines obstructing the reproducibility of published studies. In this review, we discuss the applications of AI methods and explore their benefits and limitations. We summarize the available guidelines for AI in healthcare and highlight the potential role and impact of AI models on future directions in cancer research.

The Integration of Radiomics and Artificial Intelligence in Modern Medicine

With profound effects on patient care, the role of artificial intelligence (AI) in radiomics has become a disruptive force in contemporary medicine. Radiomics, the quantitative feature extraction and analysis from medical images, offers useful imaging biomarkers that can reveal important information about the nature of diseases, how well patients respond to treatment and patient outcomes. The use of AI techniques in radiomics, such as machine learning and deep learning, has made it possible to create sophisticated computer-aided diagnostic systems, predictive models, and decision support tools. The many uses of AI in radiomics are examined in this review, encompassing its involvement of quantitative feature extraction from medical images, the machine learning, deep learning and computer-aided diagnostic (CAD) systems approaches in radiomics, and the effect of radiomics and AI on improving workflow automation and efficiency, optimize clinical trials and patient stratification. This review also covers the predictive modeling improvement by machine learning in radiomics, the multimodal integration and enhanced deep learning architectures, and the regulatory and clinical adoption considerations for radiomics-based CAD. Particular emphasis is given to the enormous potential for enhancing diagnosis precision, treatment personalization, and overall patient outcomes.

Optimized patient-specific immune checkpoint inhibitor therapies for cancer treatment based on tumor immune microenvironment modeling

Enhancing patient response to immune checkpoint inhibitors (ICIs) is crucial in cancer immunotherapy. We aim to create a data-driven mathematical model of the tumor immune microenvironment (TIME) and utilize deep reinforcement learning (DRL) to optimize patient-specific ICI therapy combined with chemotherapy (ICC). Using patients' genomic and transcriptomic data, we develop an ordinary differential equations (ODEs)-based TIME dynamic evolutionary model to characterize interactions among chemotherapy, ICIs, immune cells, and tumor cells. A DRL agent is trained to determine the personalized optimal ICC therapy. Numerical experiments with real-world data demonstrate that the proposed TIME model can predict ICI therapy response. The DRL-derived personalized ICC therapy outperforms predefined fixed schedules. For tumors with extremely low CD8 + T cell infiltration ('extremely cold tumors'), the DRL agent recommends high-dosage chemotherapy alone. For tumors with higher CD8 + T cell infiltration ('cold' and 'hot tumors'), an appropriate chemotherapy dosage induces CD8 + T cell proliferation, enhancing ICI therapy outcomes. Specifically, for 'hot tumors', chemotherapy and ICI are administered simultaneously, while for 'cold tumors', a mid-dosage of chemotherapy makes the TIME 'hotter' before ICI administration. However, in several 'cold tumors' with rapid resistant tumor cell growth, ICC eventually fails. This study highlights the potential of utilizing real-world clinical data and DRL algorithm to develop personalized optimal ICC by understanding the complex biological dynamics of a patient's TIME. Our ODE-based TIME dynamic evolutionary model offers a theoretical framework for determining the best use of ICI, and the proposed DRL agent may guide personalized ICC schedules.

Longitudinal deep neural networks for assessing metastatic brain cancer on a large open benchmark

The detection and tracking of metastatic cancer over the lifetime of a patient remains a major challenge in clinical trials and real-world care. Advances in deep learning combined with massive datasets may enable the development of tools that can address this challenge. We present NYUMets-Brain, the world's largest, longitudinal, real-world dataset of cancer consisting of the imaging, clinical follow-up, and medical management of 1,429 patients. Using this dataset we developed Segmentation-Through-Time, a deep neural network which explicitly utilizes the longitudinal structure of the data and obtained state-of-the-art results at small (<10 mm3) metastases detection and segmentation. We also demonstrate that the monthly rate of change of brain metastases over time are strongly predictive of overall survival (HR 1.27, 95%CI 1.18-1.38). We are releasing the dataset, codebase, and model weights for other cancer researchers to build upon these results and to serve as a public benchmark.

Transfer learning approach in pre-treatment CT images to predict therapeutic response in advanced malignant pleural mesothelioma

Introduction

Malignant pleural mesothelioma (MPM) is a poor-prognosis disease. Owing to the recent availability of new therapeutic options, there is a need to better assess prognosis. The initial clinical response could represent a useful parameter.

Methods

We proposed a transfer learning approach to predict an initial treatment response starting from baseline CT scans of patients with advanced/unresectable MPM undergoing first-line systemic therapy. The therapeutic response has been assessed according to the mRECIST criteria by CT scan at baseline and after two to three treatment cycles. We used three slices of baseline CT scan as input to the pre-trained convolutional neural network as a radiomic feature extractor. We identified a feature subset through a double feature selection procedure to train a binary SVM classifier to discriminate responders (partial response) from non-responders (stable or disease progression).

Results

The performance of the prediction classifiers was evaluated with an 80:20 hold-out validation scheme. We have evaluated how the developed model was robust to variations in the slices selected by the radiologist. In our dataset, 25 patients showed an initial partial response, whereas 13 patients showed progressive or stable disease. On the independent test, the proposed model achieved a median AUC and accuracy of 86.67% and 87.50%, respectively.

Conclusions

The proposed model has shown high performance even by varying the reference slices. Novel tools could help to improve the prognostic assessment of patients with MPM and to better identify subgroups of patients with different therapeutic responsiveness.

Enhancing cancer prediction in challenging screen-detected incident lung nodules using time-series deep learning

Lung cancer screening (LCS) using annual computed tomography (CT) scanning significantly reduces mortality by detecting cancerous lung nodules at an earlier stage. Deep learning algorithms can improve nodule malignancy risk stratification. However, they have typically been used to analyse single time point CT data when detecting malignant nodules on either baseline or incident CT LCS rounds. Deep learning algorithms have the greatest value in two aspects. These approaches have great potential in assessing nodule change across time-series CT scans where subtle changes may be challenging to identify using the human eye alone. Moreover, they could be targeted to detect nodules developing on incident screening rounds, where cancers are generally smaller and more challenging to detect confidently. Here, we show the performance of our Deep learning-based Computer-Aided Diagnosis model integrating Nodule and Lung imaging data with clinical Metadata Longitudinally (DeepCAD-NLM-L) for malignancy prediction. DeepCAD-NLM-L showed improved performance (AUC = 88%) against models utilizing single time-point data alone. DeepCAD-NLM-L also demonstrated comparable and complementary performance to radiologists when interpreting the most challenging nodules typically found in LCS programs. It also demonstrated similar performance to radiologists when assessed on out-of-distribution imaging dataset. The results emphasize the advantages of using time-series and multimodal analyses when interpreting malignancy risk in LCS.

Automated Lugano Metabolic Response Assessment in 18F-Fluorodeoxyglucose-Avid Non-Hodgkin Lymphoma With Deep Learning on 18F-Fluorodeoxyglucose-Positron Emission Tomography

PURPOSE

Artificial intelligence can reduce the time used by physicians on radiological assessments. For 18F-fluorodeoxyglucose-avid lymphomas, obtaining complete metabolic response (CMR) by end of treatment is prognostic.

METHODS

Here, we present a deep learning-based algorithm for fully automated treatment response assessments according to the Lugano 2014 classification. The proposed four-stage method, trained on a multicountry clinical trial (ClinicalTrials.gov identifier: NCT01287741) and tested in three independent multicenter and multicountry test sets on different non-Hodgkin lymphoma subtypes and different lines of treatment (ClinicalTrials.gov identifiers NCT02257567, NCT02500407; 20% holdout in ClinicalTrials.gov identifier NCT01287741), outputs the detected lesions at baseline and follow-up to enable focused radiologist review.

RESULTS

The method's response assessment achieved high agreement with the adjudicated radiologic responses (eg, agreement for overall response assessment of 93%, 87%, and 85% in ClinicalTrials.gov identifiers NCT01287741, NCT02500407, and NCT02257567, respectively) similar to inter-radiologist agreement and was strongly prognostic of outcomes with a trend toward higher accuracy for death risk than adjudicated radiologic responses (hazard ratio for end of treatment by-model CMR of 0.123, 0.054, and 0.205 in ClinicalTrials.gov identifiers NCT01287741, NCT02500407, and NCT02257567, compared with, respectively, 0.226, 0.292, and 0.272 for CMR by the adjudicated responses). Furthermore, a radiologist review of the algorithm's assessments was conducted. The radiologist median review time was 1.38 minutes/assessment, and no statistically significant differences were observed in the level of agreement of the radiologist with the model's response compared with the level of agreement of the radiologist with the adjudicated responses.

CONCLUSION

These results suggest that the proposed method can be incorporated into radiologic response assessment workflows in cancer imaging for significant time savings and with performance similar to trained medical experts.

Predicting histopathological features of aggressiveness in lung cancer using CT radiomics: a systematic review

PURPOSE

To examine the accuracy of CT radiomics to predict histopathological features of aggressiveness in lung cancer using a systematic review of test accuracy studies.

METHODS

Data sources searched included Medline, Embase, Web of Science, and Cochrane Library from up to 3 November 2023. Included studies reported test accuracy of CT radiomics models to detect the presence of: spread through air spaces (STAS), predominant adenocarcinoma pattern, adenocarcinoma grade, lymphovascular invasion (LVI), tumour infiltrating lymphocytes (TIL) and tumour necrosis, in patients with lung cancer. The primary outcome was test accuracy. Two reviewers independently assessed articles for inclusion and assessed methodological quality using the QUality Assessment of Diagnostic Accuracy Studies-2 tool. A single reviewer extracted data, which was checked by a second reviewer. Narrative data synthesis was performed.

RESULTS

Eleven studies were included in the final analysis. 10/11 studies were in East Asian populations. 4/11 studies investigated STAS, 6/11 investigated adenocarcinoma invasiveness or growth pattern, and 1/11 investigated LVI. No studies investigating TIL or tumour necrosis met inclusion criteria. Studies were of generally mixed to poor methodological quality. Reported accuracies for radiomic models ranged from 0.67 to 0.94.

CONCLUSION

Due to the high risk of bias and concerns regarding applicability, the evidence is inconclusive as to whether radiomic features can accurately predict prognostically important histopathological features of cancer aggressiveness. Many studies were excluded due to lack of external validation. Rigorously conducted prospective studies with sufficient external validity will be required for radiomic models to play a role in improving lung cancer outcomes.

Anatomy-specific Progression Classification in Chest Radiographs via Weakly Supervised Learning

Purpose To develop a machine learning approach for classifying disease progression in chest radiographs using weak labels automatically derived from radiology reports. Materials and Methods In this retrospective study, a twin neural network was developed to classify anatomy-specific disease progression into four categories: improved, unchanged, worsened, and new. A two-step weakly supervised learning approach was employed, pretraining the model on 243 008 frontal chest radiographs from 63 877 patients (mean age, 51.7 years ± 17.0 [SD]; 34 813 [55%] female) included in the MIMIC-CXR database and fine-tuning it on the subset with progression labels derived from consecutive studies. Model performance was evaluated for six pathologic observations on test datasets of unseen patients from the MIMIC-CXR database. Area under the receiver operating characteristic (AUC) analysis was used to evaluate classification performance. The algorithm is also capable of generating bounding-box predictions to localize areas of new progression. Recall, precision, and mean average precision were used to evaluate the new progression localization. One-tailed paired t tests were used to assess statistical significance. Results The model outperformed most baselines in progression classification, achieving macro AUC scores of 0.72 ± 0.004 for atelectasis, 0.75 ± 0.007 for consolidation, 0.76 ± 0.017 for edema, 0.81 ± 0.006 for effusion, 0.7 ± 0.032 for pneumonia, and 0.69 ± 0.01 for pneumothorax. For new observation localization, the model achieved mean average precision scores of 0.25 ± 0.03 for atelectasis, 0.34 ± 0.03 for consolidation, 0.33 ± 0.03 for edema, and 0.31 ± 0.03 for pneumothorax. Conclusion Disease progression classification models were developed on a large chest radiograph dataset, which can be used to monitor interval changes and detect new pathologic conditions on chest radiographs. Keywords: Prognosis, Unsupervised Learning, Transfer Learning, Convolutional Neural Network (CNN), Emergency Radiology, Named Entity Recognition Supplemental material is available for this article. © RSNA, 2024 See also commentary by Alves and Venkadesh in this issue.

CT-based multimodal deep learning for non-invasive overall survival prediction in advanced hepatocellular carcinoma patients treated with immunotherapy

OBJECTIVES

To develop a deep learning model combining CT scans and clinical information to predict overall survival in advanced hepatocellular carcinoma (HCC).

METHODS

This retrospective study included immunotherapy-treated advanced HCC patients from 52 multi-national in-house centers between 2018 and 2022. A multi-modal prognostic model using baseline and the first follow-up CT images and 7 clinical variables was proposed. A convolutional-recurrent neural network (CRNN) was developed to extract spatial-temporal information from automatically selected representative 2D CT slices to provide a radiological score, then fused with a Cox-based clinical score to provide the survival risk. The model's effectiveness was assessed using a time-dependent area under the receiver operating curve (AUC), and risk group stratification using the log-rank test. Prognostic performances of multi-modal inputs were compared to models of missing modality, and the size-based RECIST criteria.

RESULTS

Two-hundred seven patients (mean age, 61 years ± 12 [SD], 180 men) were included. The multi-modal CRNN model reached the AUC of 0.777 and 0.704 of 1-year overall survival predictions in the validation and test sets. The model achieved significant risk stratification in validation (hazard ratio [HR] = 3.330, p = 0.008), and test sets (HR = 2.024, p = 0.047) based on the median risk score of the training set. Models with missing modalities (the single-modal imaging-based model and the model incorporating only baseline scans) can still achieve favorable risk stratification performance (all p < 0.05, except for one, p = 0.053). Moreover, results proved the superiority of the deep learning-based model to the RECIST criteria.

CONCLUSION

Deep learning analysis of CT scans and clinical data can offer significant prognostic insights for patients with advanced HCC.

CRITICAL RELEVANCE STATEMENT

The established model can help monitor patients' disease statuses and identify those with poor prognosis at the time of first follow-up, helping clinicians make informed treatment decisions, as well as early and timely interventions.

KEY POINTS

An AI-based prognostic model was developed for advanced HCC using multi-national patients. The model extracts spatial-temporal information from CT scans and integrates it with clinical variables to prognosticate. The model demonstrated superior prognostic ability compared to the conventional size-based RECIST method.

A dual-radiomics model for overall survival prediction in early-stage NSCLC patient using pre-treatment CT images

Introduction

Radiation therapy (RT) is one of the primary treatment options for early-stage non-small cell lung cancer (ES-NSCLC). Therefore, accurately predicting the overall survival (OS) rate following radiotherapy is crucial for implementing personalized treatment strategies. This work aims to develop a dual-radiomics (DR) model to (1) predict 3-year OS in ES-NSCLC patients receiving RT using pre-treatment CT images, and (2) provide explanations between feature importanceand model prediction performance.

Methods

The publicly available TCIA Lung1 dataset with 132 ES-NSCLC patients received RT were studied: 89/43 patients in the under/over 3-year OS group. For each patient, two types of radiomic features were examined: 56 handcrafted radiomic features (HRFs) extracted within gross tumor volume, and 512 image deep features (IDFs) extracted using a pre-trained U-Net encoder. They were combined as inputs to an explainable boosting machine (EBM) model for OS prediction. The EBM's mean absolute scores for HRFs and IDFs were used as feature importance explanations. To evaluate identified feature importance, the DR model was compared with EBM using either (1) key or (2) non-key feature type only. Comparison studies with other models, including supporting vector machine (SVM) and random forest (RF), were also included. The performance was evaluated by the area under the receiver operating characteristic curve (AUCROC), accuracy, sensitivity, and specificity with a 100-fold Monte Carlo cross-validation.

Results

The DR model showed highestperformance in predicting 3-year OS (AUCROC=0.81 ± 0.04), and EBM scores suggested that IDFs showed significantly greater importance (normalized mean score=0.0019) than HRFs (score=0.0008). The comparison studies showed that EBM with key feature type (IDFs-only demonstrated comparable AUCROC results (0.81 ± 0.04), while EBM with non-key feature type (HRFs-only) showed limited AUCROC (0.64 ± 0.10). The results suggested that feature importance score identified by EBM is highly correlated with OS prediction performance. Both SVM and RF models were unable to explain key feature type while showing limited overall AUCROC=0.66 ± 0.07 and 0.77 ± 0.06, respectively. Accuracy, sensitivity, and specificity showed a similar trend.

Discussion

In conclusion, a DR model was successfully developed to predict ES-NSCLC OS based on pre-treatment CT images. The results suggested that the feature importance from DR model is highly correlated to the model prediction power.

Application of artificial intelligence in cancer diagnosis and tumor nanomedicine

Cancer is a major health concern due to its high incidence and mortality rates. Advances in cancer research, particularly in artificial intelligence (AI) and deep learning, have shown significant progress. The swift evolution of AI in healthcare, especially in tools like computer-aided diagnosis, has the potential to revolutionize early cancer detection. This technology offers improved speed, accuracy, and sensitivity, bringing a transformative impact on cancer diagnosis, treatment, and management. This paper provides a concise overview of the application of artificial intelligence in the realms of medicine and nanomedicine, with a specific emphasis on the significance and challenges associated with cancer diagnosis. It explores the pivotal role of AI in cancer diagnosis, leveraging structured, unstructured, and multimodal fusion data. Additionally, the article delves into the applications of AI in nanomedicine sensors and nano-oncology drugs. The fundamentals of deep learning and convolutional neural networks are clarified, underscoring their relevance to AI-driven cancer diagnosis. A comparative analysis is presented, highlighting the accuracy and efficiency of traditional methods juxtaposed with AI-based approaches. The discussion not only assesses the current state of AI in cancer diagnosis but also delves into the challenges faced by AI in this context. Furthermore, the article envisions the future development direction and potential application of artificial intelligence in cancer diagnosis, offering a hopeful prospect for enhanced cancer detection and improved patient prognosis.

Hybrid deep learning models for the screening of Diabetic Macular Edema in optical coherence tomography volumes

Several studies published so far used highly selective image datasets from unclear sources to train computer vision models and that may lead to overestimated results, while those studies conducted in real-life remain scarce. To avoid image selection bias, we stacked convolutional and recurrent neural networks (CNN-RNN) to analyze complete optical coherence tomography (OCT) cubes in a row and predict diabetic macular edema (DME), in a real-world diabetic retinopathy screening program. A retrospective cohort study was carried out. Throughout 4-years, 5314 OCT cubes from 4408 subjects who attended to the diabetic retinopathy (DR) screening program were included. We arranged twenty-two (22) pre-trained CNNs in parallel with a bidirectional RNN layer stacked at the bottom, allowing the model to make a prediction for the whole OCT cube. The staff of retina experts built a ground truth of DME later used to train a set of these CNN-RNN models with different configurations. For each trained CNN-RNN model, we performed threshold tuning to find the optimal cut-off point for binary classification of DME. Finally, the best models were selected according to sensitivity, specificity, and area under the receiver operating characteristics curve (AUROC) with their 95% confidence intervals (95%CI). An ensemble of the best models was also explored. 5188 cubes were non-DME and 126 were DME. Three models achieved an AUROC of 0.94. Among these, sensitivity, and specificity (95%CI) ranged from 84.1-90.5 and 89.7-93.3, respectively, at threshold 1, from 89.7-92.1 and 80-83.1 at threshold 2, and from 80.2-81 and 93.8-97, at threshold 3. The ensemble model improved these results, and lower specificity was observed among subjects with sight-threatening DR. Analysis by age, gender, or grade of DME did not vary the performance of the models. CNN-RNN models showed high diagnostic accuracy for detecting DME in a real-world setting. This engine allowed us to detect extra-foveal DMEs commonly overlooked in other studies, and showed potential for application as the first filter of non-referable patients in an outpatient center within a population-based DR screening program, otherwise ended up in specialized care.

Patient-Specific, Mechanistic Models of Tumor Growth Incorporating Artificial Intelligence and Big Data

Despite the remarkable advances in cancer diagnosis, treatment, and management over the past decade, malignant tumors remain a major public health problem. Further progress in combating cancer may be enabled by personalizing the delivery of therapies according to the predicted response for each individual patient. The design of personalized therapies requires the integration of patient-specific information with an appropriate mathematical model of tumor response. A fundamental barrier to realizing this paradigm is the current lack of a rigorous yet practical mathematical theory of tumor initiation, development, invasion, and response to therapy. We begin this review with an overview of different approaches to modeling tumor growth and treatment, including mechanistic as well as data-driven models based on big data and artificial intelligence. We then present illustrative examples of mathematical models manifesting their utility and discuss the limitations of stand-alone mechanistic and data-driven models. We then discuss the potential of mechanistic models for not only predicting but also optimizing response to therapy on a patient-specific basis. We describe current efforts and future possibilities to integrate mechanistic and data-driven models. We conclude by proposing five fundamental challenges that must be addressed to fully realize personalized care for cancer patients driven by computational models.

Three scans are better than two for follow-up: An automatic method for finding missed and misidentified lesions in cross-sectional follow-up of oncology patients

PURPOSE

Missed and misidentified neoplastic lesions in longitudinal studies of oncology patients are pervasive and may affect the evaluation of the disease status. Two newly identified patterns of lesion changes, lone lesions and non-consecutive lesion changes, may help radiologists to detect these lesions. This study evaluated a new interpretation revision workflow of lesion annotations in three or more consecutive scans based on these suspicious patterns.

METHODS

The interpretation revision workflow was evaluated on manual and computed lesion annotations in longitudinal oncology patient studies. For the manual revision, a senior radiologist and a senior neurosurgeon (the readers) manually annotated the lesions in each scan and later revised their annotations to identify missed and misidentified lesions with the workflow using the automatically detected patterns. For the computerized revision, lesion annotations were first computed with a previously trained nnU-Net and were then automatically revised with an AI-based method that automates the workflow readers' decisions. The evaluation included 67 patient studies with 2295 metastatic lesions in lung (19 patients, 83 CT scans, 1178 lesions), liver (18 patients, 77 CECT scans, 800 lesions) and brain (30 patients, 102 T1W-Gad MRI scans, 317 lesions).

RESULTS

Revision of the manual lesion annotations revealed 120 missed lesions and 20 misidentified lesions in 31 out of 67 (46%) studies. The automatic revision reduced the number of computed missed lesions by 55 and computed misidentified lesions by 164 in 51 out of 67 (76%) studies.

CONCLUSION

Automatic analysis of three or more consecutive volumetric scans helps find missed and misidentified lesions and may improve the evaluation of temporal changes of oncological lesions.

Multiparametric magnetic resonance imaging (MRI)-based radiomics model explained by the Shapley Additive exPlanations (SHAP) method for predicting complete response to neoadjuvant chemoradiotherapy in locally advanced rectal cancer: a multicenter retrospective study

Background

Predicting the response to neoadjuvant chemoradiotherapy (nCRT) before initiating treatment is essential for tailoring therapeutic strategies and monitoring prognosis in locally advanced rectal cancer (LARC). In this study, we aimed to develop and validate radiomic-based models to predict clinical and pathological complete responses (cCR and pCR, respectively) by incorporating the Shapley Additive exPlanations (SHAP) method for model interpretation.

Methods

A total of 285 patients with complete pretreatment clinical characteristics and T1-weighted (T1W) and T2-weighted (T2W) magnetic resonance imaging (MRI) at 3 centers were retrospectively recruited. The features of tumor lesions were extracted by PyRadiomics and selected using least absolute shrinkage and selection operator (LASSO) algorithm. The selected features were used to build multilayer perceptron (MLP) models alone or combined with clinical features. Area under the receiver operating characteristic curve (AUC), decision curve, and calibration curve were applied to evaluate performance of models. The SHAP method was adopted to explain the prediction models.

Results

The radiomic-based models all showed better performances than clinical models. The clinical-radiomic models showed the best differentiation on cCR and pCR with mean AUCs of 0.718 and 0.810 in the validation set, respectively. The decision curves of the clinical-radiomic models showed its values in clinical application. The SHAP method powerfully interpreted the prediction models both at a holistic and individual levels.

Conclusions

Our study highlights that the radiomic-based prediction models have more excellent abilities than clinical models and can effectively predict treatment response and optimize therapeutic strategies for patients with LARCs.

Accuracy of machine learning in the preoperative identification of ovarian borderline tumors: a meta-analysis

AIM

The objective of this study is to explore the diagnostic value of machine learning (ML) in borderline ovarian tumors through meta-analysis.

METHODS

Pubmed, Embase, Web of Science, and Cochrane Library databases were comprehensively retrieved from database inception untill February 16, 2023. The Prediction Model Risk of Bias Assessment Tool (PROBAST) was adopted to evaluate the risk of bias in the original studies. Sub-group analyses of ML were conducted according to clinical features and radiomics features. We separately discussed the discriminative value of ML for borderline vs benign and borderline vs malignant tumors.

RESULTS

Eighteen studies involving 12,778 subjects were included in our analysis. The modeling variables mainly consisted of radiomics features (n=13) and a small number of clinical features (n=5). When distinguishing between borderline and benign tumors, the ML model based on radiomic features achieved a c-index of 0.782 (95% CI: 0.732-0.831), sensitivity of 0.75 (95% CI: 0.67-0.82), and specificity of 0.75 (95% CI: 0.67-0.81) in the validation set. When distinguishing between borderline and malignant tumors, the ML model based on radiomic features achieved a c-index of 0.916 (95% CI: 0.891-0.940), sensitivity of 0.86 (95% CI: 0.78-0.91), and specificity of 0.88 (95% CI: 0.82-0.92) in the validation set. In addition, we analyzed the discriminatory ability of radiologists and found that their sensitivity was 0.26 (95% CI: 0.12-0.46) and specificity was 0.94 (95% CI: 0.90-0.97).

CONCLUSIONS

ML has tremendous potential in the preoperative diagnosis and differentiation of borderline ovarian tumors and may be more accurate than radiologists in diagnosing and differentiating borderline ovarian tumors.

Brain metastasis magnetic resonance imaging-based deep learning for predicting epidermal growth factor receptor (EGFR) mutation and subtypes in metastatic non-small cell lung cancer

Background

The preoperative identification of epidermal growth factor receptor (EGFR) mutations and subtypes based on magnetic resonance imaging (MRI) of brain metastases (BM) is necessary to facilitate individualized therapy. This study aimed to develop a deep learning model to preoperatively detect EGFR mutations and identify the location of EGFR mutations in patients with non-small cell lung cancer (NSCLC) and BM.

Methods

We included 160 and 72 patients who underwent contrast-enhanced T1-weighted (T1w-CE) and T2-weighted (T2W) MRI at Liaoning Cancer Hospital and Institute (center 1) and Shengjing Hospital of China Medical University (center 2) to form a training cohort and an external validation cohort, respectively. A multiscale feature fusion network (MSF-Net) was developed by adaptively integrating features based on different stages of residual network (ResNet) 50 and by introducing channel and spatial attention modules. The external validation set from center 2 was used to assess the performance of MSF-Net and to compare it with that of handcrafted radiomics features. Receiver operating characteristic (ROC) curves, accuracy, precision, recall, and F1-score were used to evaluate the effectiveness of the models. Gradient-weighted class activation mapping (Grad-CAM) was used to demonstrate the attention of the MSF-Net model.

Results

The developed MSF-Net generated a better diagnostic performance than did the handcrafted radiomics in terms of the microaveraged area under the curve (AUC) (MSF-Net: 0.91; radiomics: 0.80) and macroaveraged AUC (MSF-Net: 0.90; radiomics: 0.81) for predicting EGFR mutations and subtypes.

Conclusions

This study provides an end-to-end and noninvasive imaging tool for the preoperative prediction of EGFR mutation status and subtypes based on BM, which may be helpful for facilitating individualized clinical treatment plans.

An Advanced Lung Carcinoma Prediction and Risk Screening Model Using Transfer Learning

Lung cancer, also known as lung carcinoma, has a high death rate, but an early diagnosis can substantially reduce this risk. In the current era, prediction models face challenges such as low accuracy, excessive noise, and low contrast. To resolve these problems, an advanced lung carcinoma prediction and risk screening model using transfer learning is proposed. Our proposed model initially preprocesses lung computed tomography images for noise removal, contrast stretching, convex hull lung region extraction, and edge enhancement. The next phase segments the preprocessed images using the modified Bates distribution coati optimization (B-RGS) algorithm to extract key features. The PResNet classifier then categorizes the cancer as normal or abnormal. For abnormal cases, further risk screening determines whether the risk is low or high. Experimental results depict that our proposed model performs at levels similar to other state-of-the-art models, achieving enhanced accuracy, precision, and recall rates of 98.21%, 98.71%, and 97.46%, respectively. These results validate the efficiency and effectiveness of our suggested methodology in early lung carcinoma prediction and risk assessment.

Predicting treatment response in multicenter non-small cell lung cancer patients based on federated learning

BACKGROUND

Multicenter non-small cell lung cancer (NSCLC) patient data is information-rich. However, its direct integration becomes exceptionally challenging due to constraints involving different healthcare organizations and regulations. Traditional centralized machine learning methods require centralizing these sensitive medical data for training, posing risks of patient privacy leakage and data security issues. In this context, federated learning (FL) has attracted much attention as a distributed machine learning framework. It effectively addresses this contradiction by preserving data locally, conducting local model training, and aggregating model parameters. This approach enables the utilization of multicenter data with maximum benefit while ensuring privacy safeguards. Based on pre-radiotherapy planning target volume images of NSCLC patients, a multicenter treatment response prediction model is designed by FL for predicting the probability of remission of NSCLC patients. This approach ensures medical data privacy, high prediction accuracy and computing efficiency, offering valuable insights for clinical decision-making.

METHODS

We retrospectively collected CT images from 245 NSCLC patients undergoing chemotherapy and radiotherapy (CRT) in four Chinese hospitals. In a simulation environment, we compared the performance of the centralized deep learning (DL) model with that of the FL model using data from two sites. Additionally, due to the unavailability of data from one hospital, we established a real-world FL model using data from three sites. Assessments were conducted using measures such as accuracy, receiver operating characteristic curve, and confusion matrices.

RESULTS

The model's prediction performance obtained using FL methods outperforms that of traditional centralized learning methods. In the comparative experiment, the DL model achieves an AUC of 0.718/0.695, while the FL model demonstrates an AUC of 0.725/0.689, with real-world FL model achieving an AUC of 0.698/0.672.

CONCLUSIONS

We demonstrate that the performance of a FL predictive model, developed by combining convolutional neural networks (CNNs) with data from multiple medical centers, is comparable to that of a traditional DL model obtained through centralized training. It can efficiently predict CRT treatment response in NSCLC patients while preserving privacy.

EfficientNet-Based System for Detecting EGFR-Mutant Status and Predicting Prognosis of Tyrosine Kinase Inhibitors in Patients with NSCLC

We aimed to develop and validate a deep learning-based system using pre-therapy computed tomography (CT) images to detect epidermal growth factor receptor (EGFR)-mutant status in patients with non-small cell lung cancer (NSCLC) and predict the prognosis of advanced-stage patients with EGFR mutations treated with EGFR tyrosine kinase inhibitors (TKI). This retrospective, multicenter study included 485 patients with NSCLC from four hospitals. Of them, 339 patients from three centers were included in the training dataset to develop an EfficientNetV2-L-based model (EME) for predicting EGFR-mutant status, and the remaining patients were assigned to an independent test dataset. EME semantic features were extracted to construct an EME-prognostic model to stratify the prognosis of EGFR-mutant NSCLC patients receiving EGFR-TKI. A comparison of EME and radiomics was conducted. Additionally, we included patients from The Cancer Genome Atlas lung adenocarcinoma dataset with both CT images and RNA sequencing data to explore the biological associations between EME score and EGFR-related biological processes. EME obtained an area under the curve (AUC) of 0.907 (95% CI 0.840-0.926) on the test dataset, superior to the radiomics model (P = 0.007). The EME and radiomics fusion model showed better (AUC, 0.941) but not significantly increased performance (P = 0.895) compared with EME. In prognostic stratification, the EME-prognostic model achieved the best performance (C-index, 0.711). Moreover, the EME-prognostic score showed strong associations with biological pathways related to EGFR expression and EGFR-TKI efficacy. EME demonstrated a non-invasive and biologically interpretable approach to predict EGFR status, stratify survival prognosis, and correlate biological pathways in patients with NSCLC.