Breaking boundaries in radiology: redefining AI diagnostics via raw data ahead of reconstruction.

Objective.In the realm of utilizing artificial intelligence (AI) for medical image analysis, the paradigm of 'signal-image-knowledge' has remained unchanged. However, the process of 'signal to image' inevitably introduces information distortion, ultimately leading to irrecoverable biases in the 'image to knowledge' process. Our goal is to skip reconstruction and build a diagnostic model directly from the raw data (signal).Approach. This study focuses on computed tomography (CT) and its raw data (sinogram) as the research subjects. We simulate the real-world process of 'human-signal-image' using the workflow 'CT-simulated data- reconstructed CT,' and we develop a novel AI predictive model directly targeting raw data (RCTM). This model comprises orientation, spatial, and global analysis modules, embodying the fusion of local to global information extraction from raw data. We selected 1994 patients with retrospective cases of solid lung nodules and modeled different types of data.Main results. We employed predefined radiomic features to assess the diagnostic feature differences caused by reconstruction. The results indicated that approximately 14% of the features had Spearman correlation coefficients below 0.8. These findings suggest that despite the increasing maturity of CT reconstruction algorithms, they still introduce perturbations to diagnostic features. Moreover, our proposed RCTM achieved an area under the curve (AUC) of 0.863 in the diagnosis task, showcasing a comprehensive superiority over models constructed from secondary reconstructed CTs (0.840, 0.822, and 0.825). Additionally, the performance of RCTM closely resembled that of models constructed from original CT scans (0.868, 0.878, and 0.866).Significance. The diagnostic and therapeutic approach directly based on CT raw data can enhance the precision of AI models and the concept of 'signal-to-image' can be extended to other types of imaging. AI diagnostic models tailored to raw data offer the potential to disrupt the traditional paradigm of 'signal-image-knowledge', opening up new avenues for more accurate medical diagnostics.

CT-based radiomics: predicting early outcomes after percutaneous transluminal renal angioplasty in patients with severe atherosclerotic renal artery stenosis.

This study aimed to comprehensively evaluate non-contrast computed tomography (CT)-based radiomics for predicting early outcomes in patients with severe atherosclerotic renal artery stenosis (ARAS) after percutaneous transluminal renal angioplasty (PTRA). A total of 52 patients were retrospectively recruited, and their clinical characteristics and pretreatment CT images were collected. During a median follow-up period of 3.7 mo, 18 patients were confirmed to have benefited from the treatment, defined as a 20% improvement from baseline in the estimated glomerular filtration rate. A deep learning network trained via self-supervised learning was used to enhance the imaging phenotype characteristics. Radiomics features, comprising 116 handcrafted features and 78 deep learning features, were extracted from the affected renal and perirenal adipose regions. More features from the latter were correlated with early outcomes, as determined by univariate analysis, and were visually represented in radiomics heatmaps and volcano plots. After using consensus clustering and the least absolute shrinkage and selection operator method for feature selection, five machine learning models were evaluated. Logistic regression yielded the highest leave-one-out cross-validation accuracy of 0.780 (95%CI: 0.660-0.880) for the renal signature, while the support vector machine achieved 0.865 (95%CI: 0.769-0.942) for the perirenal adipose signature. SHapley Additive exPlanations was used to visually interpret the prediction mechanism, and a histogram feature and a deep learning feature were identified as the most influential factors for the renal signature and perirenal adipose signature, respectively. Multivariate analysis revealed that both signatures served as independent predictive factors. When combined, they achieved an area under the receiver operating characteristic curve of 0.888 (95%CI: 0.784-0.992), indicating that the imaging phenotypes from both regions complemented each other. In conclusion, non-contrast CT-based radiomics can be leveraged to predict the early outcomes of PTRA, thereby assisting in identifying patients with ARAS suitable for this treatment, with perirenal adipose tissue providing added predictive value.

Deep learning-based radiomics model can predict extranodal soft tissue metastasis in gastric cancer.

BACKGROUND: The potential prognostic value of extranodal soft tissue metastasis (ESTM) has been confirmed by increasing studies about gastric cancer (GC). However, the gold standard of ESTM is determined by pathologic examination after surgery, and there are no preoperative methods for assessment of ESTM yet. PURPOSE: This multicenter study aimed to develop a deep learning-based radiomics model to preoperatively identify ESTM and evaluate its prognostic value. METHODS: A total of 959 GC patients were enrolled from two centers and split into a training cohort (N = 551) and a test cohort (N = 236) for ESTM evaluation. Additionally, an external survival cohort (N = 172) was included for prognostic analysis. Four models were established based on clinical characteristics and multiphase computed tomography (CT) images for preoperative identification of ESTM, including a deep learning model, a hand-crafted radiomic model, a clinical model, and a combined model. C-index, decision curve, and calibration curve were utilized to assess the model performances. Survival analysis was conducted to explore the ability of stratifying overall survival (OS). RESULTS: The combined model showed good discrimination of the ESTM [C-indices (95% confidence interval, CI): 0.770 (0.729-0.812) and 0.761 (0.718-0.805) in training and test cohorts respectively], which outperformed deep learning model, radiomics model, and clinical model. The stratified analysis showed this model was not affected by patient's tumor size, the presence of lymphovascular invasion, and Lauren classification (p < 0.05). Moreover, the model score showed strong consistency with the OS [C-indices (95%CI): 0.723 (0.658-0.789, p < 0.0001) in the internal survival cohort and 0.715 (0.650-0.779, p < 0.0001) in the external survival cohort]. More interestingly, univariate analysis showed the model score was significantly associated with occult distant metastasis (p < 0.05) that was missed by preoperative diagnosis. CONCLUSIONS: The model combining CT images and clinical characteristics had an impressive predictive ability of both ESTM and prognosis, which has the potential to serve as an effective complement to the preoperative TNM staging system.

Radiomics and Deep Learning in Nasopharyngeal Carcinoma: A Review.

Nasopharyngeal carcinoma is a common head and neck malignancy with distinct clinical management compared to other types of cancer. Precision risk stratification and tailored therapeutic interventions are crucial to improving the survival outcomes. Artificial intelligence, including radiomics and deep learning, has exhibited considerable efficacy in various clinical tasks for nasopharyngeal carcinoma. These techniques leverage medical images and other clinical data to optimize clinical workflow and ultimately benefit patients. In this review, we provide an overview of the technical aspects and basic workflow of radiomics and deep learning in medical image analysis. We then conduct a detailed review of their applications to seven typical tasks in the clinical diagnosis and treatment of nasopharyngeal carcinoma, covering various aspects of image synthesis, lesion segmentation, diagnosis, and prognosis. The innovation and application effects of cutting-edge research are summarized. Recognizing the heterogeneity of the research field and the existing gap between research and clinical translation, potential avenues for improvement are discussed. We propose that these issues can be gradually addressed by establishing standardized large datasets, exploring the biological characteristics of features, and technological upgrades.

Comprehensive integrated analysis of MR and DCE-MR radiomics models for prognostic prediction in nasopharyngeal carcinoma.

Although prognostic prediction of nasopharyngeal carcinoma (NPC) remains a pivotal research area, the role of dynamic contrast-enhanced magnetic resonance (DCE-MR) has been less explored. This study aimed to investigate the role of DCR-MR in predicting progression-free survival (PFS) in patients with NPC using magnetic resonance (MR)- and DCE-MR-based radiomic models. A total of 434 patients with two MR scanning sequences were included. The MR- and DCE-MR-based radiomics models were developed based on 289 patients with only MR scanning sequences and 145 patients with four additional pharmacokinetic parameters (volume fraction of extravascular extracellular space (ve), volume fraction of plasma space (vp), volume transfer constant (Ktrans), and reverse reflux rate constant (kep) of DCE-MR. A combined model integrating MR and DCE-MR was constructed. Utilizing methods such as correlation analysis, least absolute shrinkage and selection operator regression, and multivariate Cox proportional hazards regression, we built the radiomics models. Finally, we calculated the net reclassification index and C-index to evaluate and compare the prognostic performance of the radiomics models. Kaplan-Meier survival curve analysis was performed to investigate the model's ability to stratify risk in patients with NPC. The integration of MR and DCE-MR radiomic features significantly enhanced prognostic prediction performance compared to MR- and DCE-MR-based models, evidenced by a test set C-index of 0.808 vs 0.729 and 0.731, respectively. The combined radiomics model improved net reclassification by 22.9%-52.6% and could significantly stratify the risk levels of patients with NPC (p = 0.036). Furthermore, the MR-based radiomic feature maps achieved similar results to the DCE-MR pharmacokinetic parameters in terms of reflecting the underlying angiogenesis information in NPC. Compared to conventional MR-based radiomics models, the combined radiomics model integrating MR and DCE-MR showed promising results in delivering more accurate prognostic predictions and provided more clinical benefits in quantifying and monitoring phenotypic changes associated with NPC prognosis.

Bilateral penile crural fractures due to traumatic injuries: A rare case report and literature review.

Penile crural fractures caused by traumatic injuries are rare urological emergencies similar to urethral bulb injuries. This case report discusses the findings of a 59-year-old patient who presented to our emergency department 12 h after an electric bicycle accident. Clinical examination revealed an elevated body temperature, bruised perineal skin, tender penis, and swollen scrotum. Imaging confirmed a penile fracture at the bilateral crus of the penis without considerable urethral trauma. The patient underwent conservative treatment and was followed up on an outpatient basis for approximately three months. Consequently, the patient is relieved of penile pain, has regained erectile function, and reports a satisfactory sexual life. Moreover, this study discusses the efficacy of conservative treatment combined with outpatient follow-up in managing bilateral penile crural fractures.

A multi-view co-training network for semi-supervised medical image-based prognostic prediction.

Prognostic prediction has long been a hotspot in disease analysis and management, and the development of image-based prognostic prediction models has significant clinical implications for current personalized treatment strategies. The main challenge in prognostic prediction is to model a regression problem based on censored observations, and semi-supervised learning has the potential to play an important role in improving the utilization efficiency of censored data. However, there are yet few effective semi-supervised paradigms to be applied. In this paper, we propose a semi-supervised co-training deep neural network incorporating a support vector regression layer for survival time estimation (Co-DeepSVS) that improves the efficiency in utilizing censored data for prognostic prediction. First, we introduce a support vector regression layer in deep neural networks to deal with censored data and directly predict survival time, and more importantly to calculate the labeling confidence of each case. Then, we apply a semi-supervised multi-view co-training framework to achieve accurate prognostic prediction, where labeling confidence estimation with prior knowledge of pseudo time is conducted for each view. Experimental results demonstrate that the proposed Co-DeepSVS has a promising prognostic ability and surpasses most widely used methods on a multi-phase CT dataset. Besides, the introduction of SVR layer makes the model more robust in the presence of follow-up bias.

Spectral CT-based nomogram for preoperative prediction of perineural invasion in locally advanced gastric cancer: a prospective study.

OBJECTIVES: This work focused on developing and validating the spectral CT-based nomogram to preoperatively predict perineural invasion (PNI) for locally advanced gastric cancer (LAGC). METHODS: This work prospectively included 196 surgically resected LAGC patients (139 males, 57 females, 59.55 ± 11.97 years) undergoing triple enhanced spectral CT scans. Patients were labeled as perineural invasion (PNI) positive and negative according to pathologic reports, then further split into primary (n = 130) and validation cohort (n = 66). We extracted clinicopathological information, follow-up data, iodine concentration (IC), and normalized IC values against to aorta (nICs) at arterial/venous/delayed phases (AP/VP/DP). Clinicopathological features and IC values between PNI positive and negative groups were compared. Multivariable logistic regression was performed to screen independent risk factors of PNI. Then, a nomogram was established, and its capability was determined by ROC curves. Its clinical use was evaluated by decision curve analysis. The correlations of PNI and the nomogram with patients' survival were explored by log-rank survival analysis. RESULTS: Borrmann classification, tumor thickness, and nICDP were independent predictors of PNI and used to build the nomogram. The nomogram yielded higher AUCs of 0.853 (0.744-0.928) and 0.782 (0.701-0.850) in primary and validation cohorts than any other parameters (p < 0.05). Both PNI and the nomogram were related to post-surgical treatment planning. Only PNI was associated with disease-free survival in the primary cohort (p < 0.05). CONCLUSION: This work prospectively established a spectral CT-based nomogram, which can effectively predict PNI preoperatively and potentially guide post-surgical treatment strategy in LAGC. KEY POINTS: • The present prospective study established a spectral CT-based nomogram for preoperative prediction of perineural invasion in LAGC. • The proposed nomogram, including morphological features and the quantitative iodine concentration values from spectral CT, had the potential to predict PNI for LAGC before surgery, along with guide post-surgical treatment planning. • Normalized iodine concentration at the delayed phase was the most valuable quantitative parameter, suggesting the importance of delayed enhancement in gastric CT.

Numerical modeling and field test of sonic crystal acoustic barriers.

The rapid development of highway traffic has gradually deteriorated the acoustic environment along the line. Sonic crystal theory provides new ideas for traffic acoustic barrier. However, the lack of practical numerical models and field test verifications has restricted the promotion and application of sonic crystal acoustic barriers (SCABs). In this study, a field test was conducted to study the noise reduction performance of SCAB. The SCAB exhibits excellent wave attenuation in the band gap, when compared with concrete acoustic barriers (CABs) along highways, the noise reduction performance in the band gap is improved by 0.5-2.1 dB(A), especially at the local peak in the highway noise spectrum. However, from the perspective of total insertion loss, CAB performs better than SCAB in all distances in the protected area. Next, the 3D FEM model is established based on the highway site and validated by the measured results. Compared with the commonly used 2D model, the 3D FEM model is more practical for considering the top diffraction and ground reflection, which influence the noise reduction performance a lot and need to be considered. To improve the noise reduction performance of SCAB, three types of optimization measures are explored. The gradient combination of scatterers can effectively improve the noise reduction effect in the low-frequency band gap, especially the high- to low-gradient layout. Besides, not only the porous sound-absorbing material but also the microperforated plates can improve the noise reduction effect, especially outside the band gap. The larger perforation rates and smaller apertures of microperforated plate are preferred in SCAB. This work provides field test support and promotes the application of SCABs in traffic noise control.

The Applications of Artificial Intelligence in Digestive System Neoplasms: A Review.

Importance: Digestive system neoplasms (DSNs) are the leading cause of cancer-related mortality with a 5-year survival rate of less than 20%. Subjective evaluation of medical images including endoscopic images, whole slide images, computed tomography images, and magnetic resonance images plays a vital role in the clinical practice of DSNs, but with limited performance and increased workload of radiologists or pathologists. The application of artificial intelligence (AI) in medical image analysis holds promise to augment the visual interpretation of medical images, which could not only automate the complicated evaluation process but also convert medical images into quantitative imaging features that associated with tumor heterogeneity. Highlights: We briefly introduce the methodology of AI for medical image analysis and then review its clinical applications including clinical auxiliary diagnosis, assessment of treatment response, and prognosis prediction on 4 typical DSNs including esophageal cancer, gastric cancer, colorectal cancer, and hepatocellular carcinoma. Conclusion: AI technology has great potential in supporting the clinical diagnosis and treatment decision-making of DSNs. Several technical issues should be overcome before its application into clinical practice of DSNs.

Artificial intelligence in gastric cancer: applications and challenges.

Gastric cancer (GC) is one of the most common malignant tumors with high mortality. Accurate diagnosis and treatment decisions for GC rely heavily on human experts' careful judgments on medical images. However, the improvement of the accuracy is hindered by imaging conditions, limited experience, objective criteria, and inter-observer discrepancies. Recently, the developments of machine learning, especially deep-learning algorithms, have been facilitating computers to extract more information from data automatically. Researchers are exploring the far-reaching applications of artificial intelligence (AI) in various clinical practices, including GC. Herein, we aim to provide a broad framework to summarize current research on AI in GC. In the screening of GC, AI can identify precancerous diseases and assist in early cancer detection with endoscopic examination and pathological confirmation. In the diagnosis of GC, AI can support tumor-node-metastasis (TNM) staging and subtype classification. For treatment decisions, AI can help with surgical margin determination and prognosis prediction. Meanwhile, current approaches are challenged by data scarcity and poor interpretability. To tackle these problems, more regulated data, unified processing procedures, and advanced algorithms are urgently needed to build more accurate and robust AI models for GC.

Semi-supervised Long-tail Endoscopic Image Classification.

Objective To explore the semi-supervised learning (SSL) algorithm for long-tail endoscopic image classification with limited annotations. Method We explored semi-supervised long-tail endoscopic image classification in HyperKvasir, the largest gastrointestinal public dataset with 23 diverse classes. Semi-supervised learning algorithm FixMatch was applied based on consistency regularization and pseudo-labeling. After splitting the training dataset and the test dataset at a ratio of 4:1, we sampled 20%, 50%, and 100% labeled training data to test the classification with limited annotations. Results The classification performance was evaluated by micro-average and macro-average evaluation metrics, with the Mathews correlation coefficient (MCC) as the overall evaluation. SSL algorithm improved the classification performance, with MCC increasing from 0.8761 to 0.8850, from 0.8983 to 0.8994, and from 0.9075 to 0.9095 with 20%, 50%, and 100% ratio of labeled training data, respectively. With a 20% ratio of labeled training data, SSL improved both the micro-average and macro-average classification performance; while for the ratio of 50% and 100%, SSL improved the micro-average performance but hurt macro-average performance. Through analyzing the confusion matrix and labeling bias in each class, we found that the pseudo-based SSL algorithm exacerbated the classifier's preference for the head class, resulting in improved performance in the head class and degenerated performance in the tail class. Conclusion SSL can improve the classification performance for semi-supervised long-tail endoscopic image classification, especially when the labeled data is extremely limited, which may benefit the building of assisted diagnosis systems for low-volume hospitals. However, the pseudo-labeling strategy may amplify the effect of class imbalance, which hurts the classification performance for the tail class.

Deep learning for predicting immunotherapeutic efficacy in advanced non-small cell lung cancer patients: a retrospective study combining progression-free survival risk and overall survival risk.

Background: Radiomics based on computed tomography (CT) images is potential in promoting individualized treatment of non-small cell lung cancer (NSCLC), however, its role in immunotherapy needs further exploration. The aim of this study was to develop a CT-based radiomics score to predict the efficacy of immune checkpoint inhibitor (ICI) monotherapy in patients with advanced NSCLC. Methods: Two hundred and thirty-six ICI-treated patients were retrospectively included and divided into a training cohort (n=188) and testing cohort (n=48) at a ratio of 8 to 2. The efficacy outcomes of ICI were evaluated based on overall survival (OS) and progression-free survival (PFS). We designed a survival network and combined it with a Cox regression model to obtain patients' OS risk score (OSRS) and PFS risk score (PFSRS). Results: Based on OSRS and PFSRS, patients were divided into high- and low-risk groups in the training cohort and the test cohort with distinctly different [training cohort, log-rank P<0.001, hazard ratio (HR): 4.14; test cohort, log-rank P=0.014, HR: 4.54] and PFS (training cohort, log-rank P<0.001, HR: 4.52; test cohort, log-rank P<0.001, HR: 6.64). Further joint evaluation of OSRS and PFSRS showed that both were significant in the Cox regression model (P<0.001), and multi-overall survival risk score (MOSRS) displayed more outstanding stratification capabilities than OSRS in both the training (P<0.001) and test cohorts (P=0.002). None of the clinical characteristics were significant in the Cox regression model, and the score that predicted the best immune response was not as good as the risk score from follow-up information in the performance of prognostic stratification. Conclusions: We developed a CT imaging-based score with the potential to become an independent prognostic factor to screen patients who would benefit from ICI treatment, which suggested that CT radiomics could be applied for individualized immunotherapy of NSCLC. Our findings should be further validated by future larger multicenter study.

The potential of prostate gland radiomic features in identifying the Gleason score.

BACKGROUND: Gleason score (GS) is one of the most critical predictors of diagnosing prostate cancer (PCa). The prostate gland, including both lesions and their microenvironment, may contain more comprehensive information about the PCa. We aimed to investigate the potential of prostate gland radiomic features in identifying Gleason scores (GS) < 7, = 7, and >7. METHODS: We retrospectively examined preoperative magnetic resonance imaging (MRI) results, clinical data, and postoperative pathological findings from 489 PCa patients. The three-dimensional (3D) and two-dimensional (2D) radiomic features were extracted from the manually segmented 3D prostate gland and its maximum 2D layer on MRI, respectively. Significant features were selected, and sequence signatures were then developed via multi-class linear regression (MLR) accordingly. Subsequently, 2D and 3D radiomic models were constructed by applying MLR to the combination of the sequence signatures, respectively. The stability of the significant features was discussed by their average ranking in the other 30 random cohorts. Based on our distance matrix algorithm, we generated different regions of interest to simulate the manual segmentation biases and discuss the model's tolerance to them. RESULTS: Our 2D model reached a C-index of 0.728 and an average area under the receiver operating characteristic curve of 0.794 in the validation cohort. The corresponding key features were stable, with an average ranking of the top 8.352% in 30 random cohorts, and the model could tolerate a segmentation boundary deviation of 2 mm without significant performance degradation. CONCLUSION: 2D prostate-gland-MRI-based radiomic features showed stable potential in identifying GS.

Deep learning-based AI model for signet-ring cell carcinoma diagnosis and chemotherapy response prediction in gastric cancer.

PURPOSE: We aimed to develop a noninvasive artificial intelligence (AI) model to diagnose signet-ring cell carcinoma (SRCC) of gastric cancer (GC) and identify patients with SRCC who could benefit from postoperative chemotherapy based on preoperative contrast-enhanced computed tomography (CT). METHODS: A total of 855 GC patients with 855 single GCs were included, of which 249 patients were diagnosed as SRCC by histopathologic examinations. The AI model was generated with clinical, handcrafted radiomic, and deep learning features. Model diagnostic performance was measured by area under the receiver operating characteristic curve (AUC), sensitivity, and specificity, while predictive performance was measured by Kaplan-Meier curves. RESULTS: In the test cohort (n = 257), the AUC, sensitivity, and specificity of our AI model for diagnosing SRCC were 0.786 (95% CI: 0.721-0.845), 77.3%, and 69.2%, respectively. For the entire cohort, patients with AI-predicted high risk had a significantly shorter median OS compared with those with low risk (median overall survival [OS], 38.8 vs. 64.2 months, p = 0.009). Importantly, in pathologically confirmed advanced SRCC patients, AI-predicted high-risk status was indicative of a shorter overall survival (median overall survival [OS], 31.0 vs. 54.4 months, p = 0.036) and marked chemotherapy resistance, whereas AI-predicted low-risk status had substantial chemotherapy benefit (median OS [without vs. with chemotherapy], 26.0 vs. not reached, p = 0.013). CONCLUSIONS: The CT-based AI model demonstrated good performance for diagnosing SRCC, stratifying patient prognosis, and predicting chemotherapy responses. Advanced SRCC patients with AI-predicted low-risk status may benefit substantially from adjuvant chemotherapy.

A deep learning-based radiomic nomogram for prognosis and treatment decision in advanced nasopharyngeal carcinoma: A multicentre study.

BACKGROUND: Induction chemotherapy (ICT) plus concurrent chemoradiotherapy (CCRT) and CCRT alone were the optional treatment regimens in locoregionally advanced nasopharyngeal carcinoma (NPC) patients. Currently, the choice of them remains equivocal in clinical practice. We aimed to develop a deep learning-based model for treatment decision in NPC. METHODS: A total of 1872 patients with stage T3N1M0 NPC were enrolled from four Chinese centres and received either ICT+CCRT or CCRT. A nomogram was constructed for predicting the prognosis of patients with different treatment regimens using multi-task deep learning radiomics and pre-treatment MR images, based on which an optimal treatment regimen was recommended. Model performance was assessed by the concordance index (C-index) and the Kaplan-Meier estimator. FINDINGS: The nomogram showed excellent prognostic ability for disease-free survival in both the CCRT (C-index range: 0.888-0.921) and ICT+CCRT (C-index range: 0.784-0.830) groups. According to the prognostic difference between treatments using the nomogram, patients were divided into the ICT-preferred and CCRT-preferred groups. In the ICT-preferred group, patients receiving ICT+CCRT exhibited prolonged survival over those receiving CCRT in the internal and external test cohorts (hazard ratio [HR]: 0.17, p<0.001 and 0.24, p=0.02); while the trend was opposite in the CCRT-preferred group (HR: 6.24, p<0.001 and 12.08, p<0.001). Similar results for treatment decision using the nomogram were obtained in different subgroups stratified by clinical factors and MR acquisition parameters. INTERPRETATION: Our nomogram could predict the prognosis of T3N1M0 NPC patients with different treatment regimens and accordingly recommend an optimal treatment regimen, which may serve as a potential tool for promoting personalized treatment of NPC. FUNDING: National Key R&D Program of China, National Natural Science Foundation of China, Beijing Natural Science Foundation, Strategic Priority Research Program of CAS, Project of High-Level Talents Team Introduction in Zhuhai City, Beijing Natural Science Foundation, Beijing Nova Program, Youth Innovation Promotion Association CAS.

Integrating No.3 lymph nodes and primary tumor radiomics to predict lymph node metastasis in T1-2 gastric cancer.

BACKGROUND: This study aimed to develope and validate a radiomics nomogram by integrating the quantitative radiomics characteristics of No.3 lymph nodes (LNs) and primary tumors to better predict preoperative lymph node metastasis (LNM) in T1-2 gastric cancer (GC) patients. METHODS: A total of 159 T1-2 GC patients who had undergone surgery with lymphadenectomy between March 2012 and November 2017 were retrospectively collected and divided into a training cohort (n = 80) and a testing cohort (n = 79). Radiomic features were extracted from both tumor region and No. 3 station LNs based on computed tomography (CT) images per patient. Then, key features were selected using minimum redundancy maximum relevance algorithm and fed into two radiomic signatures, respectively. Meanwhile, the predictive performance of clinical risk factors was studied. Finally, a nomogram was built by merging radiomic signatures and clinical risk factors and evaluated by the area under the receiver operator characteristic curve (AUC) as well as decision curve. RESULTS: Two radiomic signatures, reflecting phenotypes of the tumor and LNs respectively, were significantly associated with LN metastasis. A nomogram incorporating two radiomic signatures and CT-reported LN metastasis status showed good discrimination of LN metastasis in both the training cohort (AUC 0.915; 95% confidence interval [CI] 0.832-0.998) and testing cohort (AUC 0.908; 95% CI 0.814-1.000). The decision curve also indicated its potential clinical usefulness. CONCLUSIONS: The nomogram received favorable predictive accuracy in predicting No.3 LNM in T1-2 GC, and the nomogram showed positive role in predicting LNM in No.4 LNs. The nomogram may be used to predict LNM in T1-2 GC and could assist the choice of therapy.