Deep Convolutional Neural Network Based on Computed Tomography Images for the Preoperative Diagnosis of Occult Peritoneal Metastasis in Advanced Gastric Cancer

Dual-energy CT quantitative parameters to evaluate occult peritoneal metastasis in advanced gastric cancer preoperatively

PURPOSE

To explore whether dual-energy CT (DECT) quantitative parameters could provide analytic value for the diagnosis of patients with occult peritoneal metastasis (OPM) in advanced gastric cancer preoperatively.

MATERIALS AND METHODS

This retrospective study included 219 patients with advanced gastric cancer and DECT scans. The patient's clinical data and DECT related iodine concentration (IC) parameters and effective atomic number (Zeff) were collated and analyzed among noun-peritoneal metastasis (NPM), OPM and radiologically peritoneal metastasis (RPM) groups. The predictive performance of the DECT parameters was compared with that of the conventional CT features and clinical characteristics through evaluating area under curve of the precision-recall (AUC-PR), F1 score, balanced accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).

RESULTS

Borrmann IV type diagnosed on CT and serum tumor indicator CA125 index were statistically different between the NPM and OPM groups. DECT parameters included IC, normalized IC (NIC), and Zeff of PM group were lower than the NPM group. The DECT predictive nomogram combined three independent DECT parameters produced a better diagnostic performance than the conventional CT feature Borrmann IV type and serum CA125 index in AUC-PR with 0.884 vs 0.368 vs 0.189, but similar to the combined indicator which was based on the DECT parameters, the conventional CT feature, and serum CA125 index in AUC-PR with 0.884 vs 0.918.

CONCLUSION

The lower quantitative NIC, IC ratio, and Zeff on DECT was associated with peritoneal metastasis in advanced gastric cancer and was promising to identify patients with OPM noninvasively.

Application of artificial intelligence in the diagnosis, treatment, and recurrence prediction of peritoneal carcinomatosis

Peritoneal carcinomatosis (PC) is a type of secondary cancer which is not sensitive to conventional intravenous chemotherapy. Treatment strategies for PC are usually palliative rather than curative. Recently, artificial intelligence (AI) has been widely used in the medical field, making the early diagnosis, individualized treatment, and accurate prognostic evaluation of various cancers, including mediastinal malignancies, colorectal cancer, lung cancer more feasible. As a branch of computer science, AI specializes in image recognition, speech recognition, automatic large-scale data extraction and output. AI technologies have also made breakthrough progress in the field of peritoneal carcinomatosis (PC) based on its powerful learning capacity and efficient computational power. AI has been successfully applied in various approaches in PC diagnosis, including imaging, blood tests, proteomics, and pathological diagnosis. Due to the automatic extraction function of the convolutional neural network and the learning model based on machine learning algorithms, AI-assisted diagnosis types are associated with a higher accuracy rate compared to conventional diagnosis methods. In addition, AI is also used in the treatment of peritoneal cancer, including surgical resection, intraperitoneal chemotherapy, systemic chemotherapy, which significantly improves the survival of patients with PC. In particular, the recurrence prediction and emotion evaluation of PC patients are also combined with AI technology, further improving the quality of life of patients. Here we have comprehensively reviewed and summarized the latest developments in the application of AI in PC, helping oncologists to comprehensively diagnose PC and provide more precise treatment strategies for patients with PC.

Utility of Combined Use of Imaging Features From Abdominopelvic CT and CA 125 to Identify Presence of CT Occult Peritoneal Metastases in Advanced Gastric Cancer

OBJECTIVE

The purpose of this study is to identify the presence of occult peritoneal metastasis (OPM) in patients with advanced gastric cancer (AGC) by using clinical characteristics and abdominopelvic computed tomography (CT) features.

METHODS

This retrospective study included 66 patients with OPM and 111 patients without peritoneal metastasis (non-PM [NPM]) who underwent preoperative contrast-enhanced CT between January 2020 and December 2021. Occult PMs means PMs that are missed by CT but later diagnosed by laparoscopy or laparotomy. Patients with NPM means patients have neither PM nor other distant metastases, indicating there is no evidence of distant metastases in patients with AGC. Patients' clinical characteristics and CT features such as tumor marker, Borrmann IV, enhancement patterns, and pelvic ascites were observed by 2 experienced radiologists. Computed tomography features and clinical characteristics were combined to construct an indicator for identifying the presence of OPM in patients with AGC based on a logistic regression model. Receiver operating characteristic curves and the area under the receiver operating characteristic curve (AUC) were generated to assess the diagnostic performance of the combined indicator.

RESULTS

Four independent predictors (Borrmann IV, pelvic ascites, carbohydrate antigen 125, and normalized arterial CT value) differed significantly between OPM and NPM and performed outstandingly in distinguishing patients with OPM from those without PM (AUC = 0.643-0.696). The combined indicator showed a higher AUC value than the independent risk factors (0.820 vs 0.643-0.696).

CONCLUSIONS

The combined indicator based on abdominopelvic CT features and carbohydrate antigen 125 may assist clinicians in identifying the presence of CT OPMs in patients with AGC.

Predictive performance of radiomics for peritoneal metastasis in patients with gastric cancer: a meta-analysis and radiomics quality assessment

PURPOSE

The purpose of this meta-analysis is to systematically review the diagnostic performance of radiomic techniques in predicting peritoneal metastasis in patients with gastric cancer, and to evaluate the quality of current research.

METHODS

We searched PubMed, Web of Science, EBSCO, Embase, and Cochrane databases for relevant studies up to April 3, 2023. Data extraction and quality evaluation were performed by two independent reviewers. Then we performed statistical analysis, including plotting the forest plot and summary receiver operating characteristic (SROC) curve, and source of heterogeneity analysis, through the MIDAS module in Stata 15. We performed meta-regression and subgroup analyses to analyze the sources of heterogeneity. Using the QUADAS-2 scale and the RQS scale to assess the quality of retrieved studies.

RESULTS

Ten studies with 6199 patients were finally included in our meta-analysis. Pooled sensitivity and specificity were 0.77 (95% confidence interval [CI]: 0.66, 0.86), and 0.88 (95% CI 0.80, 0.93), respectively. The overall AUC was 0.89 (95% CI 0.86, 0.92). The heterogeneity of this meta-analysis was high, with I2 = 88% (95% CI 75,100). The result of meta-regression showed that QUADAS-2 results, RQS results and machine learning method led to heterogeneity in sensitivity and specificity (P < 0.05). Furthermore, the image segmentation area and the presence or absence of combined clinical factors were associated with sensitivity heterogeneity and specificity heterogeneity, respectively.

CONCLUSION

Undoubtedly, radiomics has potential value in diagnosing peritoneal metastasis of gastric cancer, but the quality of current research is inconsistent, and more standardized and high-quality research is still needed in the future to achieve the transformation of radiomics results into clinical applications.

Artificial intelligence applications in computed tomography in gastric cancer: a narrative review

Background and Objective

Artificial intelligence (AI) is a revolutionary technique which is deeply impacting and reshaping clinical practice in oncology. This review aims to summarize the current status of the clinical application of AI-based computed tomography (CT) for gastric cancer (GC), focusing on diagnosis, genetic status detection and risk prediction of metastasis, prognosis and treatment efficacy. The challenges and prospects for future research will also be discussed.

Methods

We searched the PubMed/MEDLINE database to identify clinical studies published between 1990 and November 2022 that investigated AI applications in CT in GC. The major findings of the verified studies were summarized.

Key Content and Findings

AI applications in CT images have attracted considerable attention in various fields such as diagnosis, prediction of metastasis risk, survival, and treatment response. These emerging techniques have shown a high potential to outperform clinicians in diagnostic accuracy and time-saving.

Conclusions

AI-powered tools showed great potential to increase diagnostic accuracy and reduce radiologists' workload. However, the goal of AI is not to replace human ability but to help oncologists make decisions in their practice. Therefore, radiologists should play a predominant role in AI applications and decide the best ways to integrate these complementary techniques within clinical practice.

Artificial Intelligence in Gastric Cancer Imaging With Emphasis on Diagnostic Imaging and Body Morphometry

Gastric cancer remains a significant global health concern, coercing the need for advancements in imaging techniques for ensuring accurate diagnosis and effective treatment planning. Artificial intelligence (AI) has emerged as a potent tool for gastric-cancer imaging, particularly for diagnostic imaging and body morphometry. This review article offers a comprehensive overview of the recent developments and applications of AI in gastric cancer imaging. We investigated the role of AI imaging in gastric cancer diagnosis and staging, showcasing its potential to enhance the accuracy and efficiency of these crucial aspects of patient management. Additionally, we explored the application of AI body morphometry specifically for assessing the clinical impact of gastrectomy. This aspect of AI utilization holds significant promise for understanding postoperative changes and optimizing patient outcomes. Furthermore, we examine the current state of AI techniques for the prognosis of patients with gastric cancer. These prognostic models leverage AI algorithms to predict long-term survival outcomes and assist clinicians in making informed treatment decisions. However, the implementation of AI techniques for gastric cancer imaging has several limitations. As AI continues to evolve, we hope to witness the translation of cutting-edge technologies into routine clinical practice, ultimately improving patient care and outcomes in the fight against gastric cancer.

The Role of Staging Laparoscopy for Gastric Cancer Patients: Current Evidence and Future Perspectives

A significant proportion of patients diagnosed with gastric cancer is discovered with peritoneal metastases at laparotomy. Despite the continuous improvement in the performance of radiological imaging, the preoperative recognition of such an advanced disease is still challenging during the diagnostic work-up, since the sensitivity of CT scans to peritoneal carcinomatosis is not always adequate. Staging laparoscopy offers the chance to significantly increase the rate of promptly diagnosed peritoneal metastases, thus reducing the number of unnecessary laparotomies and modifying the initial treatment strategy of gastric cancer. The aim of this review was to provide a comprehensive summary of the current literature regarding the role of staging laparoscopy in the management of gastric cancer. Indications, techniques, accuracy, advantages, and limitations of staging laparoscopy and peritoneal cytology were discussed. Furthermore, a focus on current evidence regarding the application of artificial intelligence and image-guided surgery in staging laparoscopy was included in order to provide a picture of the future perspectives of this technique and its integration with modern tools in the preoperative management of gastric cancer.

Deep Learning in the Detection of Rare Fractures - Development of a "Deep Learning Convolutional Network" Model for Detecting Acetabular Fractures

BACKGROUND

Fracture detection by artificial intelligence and especially Deep Convolutional Neural Networks (DCNN) is a topic of growing interest in current orthopaedic and radiological research. As learning a DCNN usually needs a large amount of training data, mostly frequent fractures as well as conventional X-ray are used. Therefore, less common fractures like acetabular fractures (AF) are underrepresented in the literature. The aim of this pilot study was to establish a DCNN for detection of AF using computer tomography (CT) scans.

METHODS

Patients with an acetabular fracture were identified from the monocentric consecutive pelvic injury registry at the BG Trauma Center XXX from 01/2003 - 12/2019. All patients with unilateral AF and CT scans available in DICOM-format were included for further processing. All datasets were automatically anonymised and digitally post-processed. Extraction of the relevant region of interests was performed and the technique of data augmentation (DA) was implemented to artificially increase the number of training samples. A DCNN based on Med3D was used for autonomous fracture detection, using global average pooling (GAP) to reduce overfitting.

RESULTS

From a total of 2,340 patients with a pelvic fracture, 654 patients suffered from an AF. After screening and post-processing of the datasets, a total of 159 datasets were enrolled for training of the algorithm. A random assignment into training datasets (80%) and test datasets (20%) was performed. The technique of bone area extraction, DA and GAP increased the accuracy of fracture detection from 58.8% (native DCNN) up to an accuracy of 82.8% despite the low number of datasets.

CONCLUSION

The accuracy of fracture detection of our trained DCNN is comparable to published values despite the low number of training datasets. The techniques of bone extraction, DA and GAP are useful for increasing the detection rates of rare fractures by a DCNN. Based on the used DCNN in combination with the described techniques from this pilot study, the possibility of an automatic fracture classification of AF is under investigation in a multicentre study.

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.

Transformer-Based Deep-Learning Algorithm for Discriminating Demyelinating Diseases of the Central Nervous System With Neuroimaging

Background

Differential diagnosis of demyelinating diseases of the central nervous system is a challenging task that is prone to errors and inconsistent reading, requiring expertise and additional examination approaches. Advancements in deep-learning-based image interpretations allow for prompt and automated analyses of conventional magnetic resonance imaging (MRI), which can be utilized in classifying multi-sequence MRI, and thus may help in subsequent treatment referral.

Methods

Imaging and clinical data from 290 patients diagnosed with demyelinating diseases from August 2013 to October 2021 were included for analysis, including 67 patients with multiple sclerosis (MS), 162 patients with aquaporin 4 antibody-positive (AQP4+) neuromyelitis optica spectrum disorder (NMOSD), and 61 patients with myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD). Considering the heterogeneous nature of lesion size and distribution in demyelinating diseases, multi-modal MRI of brain and/or spinal cord were utilized to build the deep-learning model. This novel transformer-based deep-learning model architecture was designed to be versatile in handling with multiple image sequences (coronal T2-weighted and sagittal T2-fluid attenuation inversion recovery) and scanning locations (brain and spinal cord) for differentiating among MS, NMOSD, and MOGAD. Model performances were evaluated using the area under the receiver operating curve (AUC) and the confusion matrices measurements. The classification accuracy between the fusion model and the neuroradiological raters was also compared.

Results

The fusion model that was trained with combined brain and spinal cord MRI achieved an overall improved performance, with the AUC of 0.933 (95%CI: 0.848, 0.991), 0.942 (95%CI: 0.879, 0.987) and 0.803 (95%CI: 0.629, 0.949) for MS, AQP4+ NMOSD, and MOGAD, respectively. This exceeded the performance using the brain or spinal cord MRI alone for the identification of the AQP4+ NMOSD (AUC of 0.940, brain only and 0.689, spinal cord only) and MOGAD (0.782, brain only and 0.714, spinal cord only). In the multi-category classification, the fusion model had an accuracy of 81.4%, which was significantly higher compared to rater 1 (64.4%, p=0.04<0.05) and comparable to rater 2 (74.6%, p=0.388).

Conclusion

The proposed novel transformer-based model showed desirable performance in the differentiation of MS, AQP4+ NMOSD, and MOGAD on brain and spinal cord MRI, which is comparable to that of neuroradiologists. Our model is thus applicable for interpretating conventional MRI in the differential diagnosis of demyelinating diseases with overlapping lesions.

Mass Deployment of Deep Neural Network: Real-Time Proof of Concept With Screening of Intracranial Hemorrhage Using an Open Data Set

BACKGROUND

Intracranial hemorrhage (ICH) is considered an emergency that requires rapid medical or surgical management. Previous studies have used artificial intelligence to attempt to expedite the diagnosis of this pathology on neuroimaging. However, these studies have used local, institution-specific data for training of networks that limit deployment of across broader hospital networks or regions because of data biases.

OBJECTIVE

To demonstrate the creation of a neural network based on an openly available imaging data tested on data from our institution demonstrating a high-efficacy, institution-agnostic network.

METHODS

A data set was created from publicly available noncontrast computed tomography images of known ICH. These data were used to train a neural network using distinct windowing and augmentation. This network was then validated in 2 phases using cohort-based (phase 1) and longitudinal (phase 2) approaches.

RESULTS

Our convolutional neural network was trained on 752 807 openly available slices, which included 112 762 slices containing intracranial hemorrhage. In phase 1, the final network performance for intracranial hemorrhage showed a receiver operating characteristic curve (AUC) of 0.99. At the inflection point, our model showed a sensitivity of 98% at a threshold specificity of 99%. In phase 2, we obtained an AUC of 0.98 after analysis of 726 scans with a negative predictive value of 99.70% (n = 726).

CONCLUSION

We demonstrate an effective neural network trained on completely open data for screening ICH at an unrelated institution. This study demonstrates a proof of concept for screening networks for multiple sites while maintaining high efficacy.

Safety and Efficacy of Intraperitoneal Paclitaxel Plus Intravenous Fluorouracil, Leucovorin, and Oxaliplatin (FOLFOX) for Gastric Cancer with Peritoneal Metastasis

BACKGROUND

Peritoneal metastasis (PM) remains a major obstacle in the treatment of stage IV gastric cancer. This is a dose-escalation study of intraperitoneal (IP) paclitaxel combined with intravenous (IV) fluorouracil, leucovorin, and oxaliplatin (FOLFOX) to determine the recommended phase II dose in gastric cancer patients.

METHODS

Patients with gastric adenocarcinoma and PM were enrolled. The recommended phase II dose of IP paclitaxel was determined using the standard "3 + 3" dose escalation with planned doses ranging from 40 to 100 mg/m². IV FOLFOX was administered on the same day (oxaliplatin 100 mg/m² (day 1), leucovorin 100 mg/m² (day 1), fluorouracil 2,400 mg/m² over 46 hours (day 1)). Both IP and IV regimens were repeated every 2 weeks.

RESULTS

Among the 13 patients, there was no DLT at 40 and 60 mg/m². Two patients had grade 3 febrile neutropenia at 80 mg/m², and the recommended phase II dose was 60 mg/m². Other patients underwent IP paclitaxel and FOLFOX without serious adverse events. Seven patients underwent second-look diagnostic laparoscopy, and the average change in PCI score was -7.0 ± 9.7. Conversion surgery rate was 23.1% (n = 3). The median overall survival was 16.6 months (95% confidence interval, 16.6-N/A), and progression-free survival was 9.6 months (95% confidence interval, 4.7-N/A). All adverse events were tolerable and manageable.

CONCLUSIONS

The biweekly regimen of IP paclitaxel and FOLFOX is safe and the recommended dose of IP paclitaxel for a phase II trial is 60 mg/m².

Comparison of clinical-computed tomography model with 2D and 3D radiomics models to predict occult peritoneal metastases in advanced gastric cancer

PURPOSE

To compare the ability of a clinical-computed tomography (CT) model vs. 2D and 3D radiomics models for predicting occult peritoneal metastasis (PM) in patients with advanced gastric cancer (AGC).

METHODS

In this retrospective study, we included 49 patients with occult PM and 49 control patients (without PM) who underwent preoperative CT and subsequent surgery between January 2016 and December 2018. Clinical information and CT semantic features were collected, and CT radiomics features were extracted. A predictive clinical-CT model was created using multivariate logistic regression. The least absolute shrinkage and selection operator algorithm and logistic regression were used for constructing 2D and 3D radiomics models. These models were validated with an external cohort (n = 30). Receiver operating characteristics curve with area under the curve (AUC), sensitivity, and specificity were used to evaluate predictive performance.

RESULTS

Tumor size, mild ascites, and serum CA125 were independent factors predictive of occult PM. The clinical-CT model of these independent factors showed better diagnostic performance than 2D and 3D radiomics models. In the external validation cohort, the AUCs of different models were as follows-clinical-CT model: 0.853 (sensitivity, 66.7%; specificity, 93.3%); 2D radiomics model: 0.622 (sensitivity, 80.0%; specificity, 46.7%); and 3D radiomics model: 0.676 (sensitivity, 60.0%; specificity, 86.0%). The clinical-CT model nomogram showed good clinical predictive efficiency to assess occult PM.

CONCLUSION

The clinical-CT model was better than the radiomics models in predicting occult PM in AGC.

A Bounding Box-Based Radiomics Model for Detecting Occult Peritoneal Metastasis in Advanced Gastric Cancer: A Multicenter Study

Purpose

To develop a bounding box (BBOX)-based radiomics model for the preoperative diagnosis of occult peritoneal metastasis (OPM) in advanced gastric cancer (AGC) patients.

Materials and Methods

599 AGC patients from 3 centers were retrospectively enrolled and were divided into training, validation, and testing cohorts. The minimum circumscribed rectangle of the ROIs for the largest tumor area (R_BBOX), the nonoverlapping area between the tumor and R_BBOX (peritumoral area; PERI) and the smallest rectangle that could completely contain the tumor determined by a radiologist (M_BBOX) were used as inputs to extract radiomic features. Multivariate logistic regression was used to construct a radiomics model to estimate the preoperative probability of OPM in AGC patients.

Results

The M_BBOX model was not significantly different from R_BBOX in the validation cohort [AUC: M_BBOX model 0.871 (95% CI, 0.814–0.940) vs. R_BBOX model 0.873 (95% CI, 0.820–0.940); p = 0.937]. M_BBOX was selected as the final radiomics model because of its extremely low annotation cost and superior OPM discrimination performance (sensitivity of 85.7% and specificity of 82.8%) over the clinical model, and this radiomics model showed comparable diagnostic efficacy in the testing cohort.

Conclusions

The BBOX-based radiomics could serve as a simpler reliable and powerful tool for the preoperative diagnosis of OPM in AGC patients. And M_BBOX-based radiomics is simpler and less time consuming.

Machine Learning for Future Subtyping of the Tumor Microenvironment of Gastro-Esophageal Adenocarcinomas

Tumor progression involves an intricate interplay between malignant cells and their surrounding tumor microenvironment (TME) at specific sites. The TME is dynamic and is composed of stromal, parenchymal, and immune cells, which mediate cancer progression and therapy resistance. Evidence from preclinical and clinical studies revealed that TME targeting and reprogramming can be a promising approach to achieve anti-tumor effects in several cancers, including in GEA. Thus, it is of great interest to use modern technology to understand the relevant components of programming the TME. Here, we discuss the approach of machine learning, which recently gained increasing interest recently because of its ability to measure tumor parameters at the cellular level, reveal global features of relevance, and generate prognostic models. In this review, we discuss the relevant stromal composition of the TME in GEAs and discuss how they could be integrated. We also review the current progress in the application of machine learning in different medical disciplines that are relevant for the management and study of GEA.

Dual-Energy Computed Tomography-Based Radiomics to Predict Peritoneal Metastasis in Gastric Cancer

Objective

To develop and validate a dual-energy computed tomography (DECT) derived radiomics model to predict peritoneal metastasis (PM) in patients with gastric cancer (GC).

Methods

This retrospective study recruited 239 GC (non-PM = 174, PM = 65) patients with histopathological confirmation for peritoneal status from January 2015 to December 2019. All patients were randomly divided into a training cohort (n = 160) and a testing cohort (n = 79). Standardized iodine-uptake (IU) images and 120-kV-equivalent mixed images (simulating conventional CT images) from portal-venous and delayed phases were used for analysis. Two regions of interest (ROIs) including the peritoneal area and the primary tumor were independently delineated. Subsequently, 1691 and 1226 radiomics features were extracted from the peritoneal area and the primary tumor from IU and mixed images on each phase. Boruta and Spearman correlation analysis were used for feature selection. Three radiomics models were established, including the R_IU model for IU images, the R_MIX model for mixed images and the combined radiomics model (the R_comb model). Random forest was used to tune the optimal radiomics model. The performance of the clinical model and human experts to assess PM was also recorded.

Results

Fourteen and three radiomics features with low redundancy and high importance were extracted from the IU and mixed images, respectively. The R_IU model showed significantly better performance to predict PM than the R_MIX model in the training cohort (AUC, 0.981 vs. 0.917, p = 0.034). No improvement was observed in the R_comb model (AUC = 0.967). The R_IU model was the optimal radiomics model which showed no overfitting in the testing cohort (AUC = 0.967, p = 0.528). The R_IU model demonstrated significantly higher predictive value on peritoneal status than the clinical model and human experts in the testing cohort (AUC, 0.785, p = 0.005; AUC, 0.732, p <0.001, respectively).

Conclusion

DECT derived radiomics could serve as a non-invasive and easy-to-use biomarker to preoperatively predict PM for GC, providing opportunity for those patients to tailor appropriate treatment.