Multimodal lung tumor image recognition algorithm based on integrated convolutional neural network

BackgroundTo distinguish early‐stage lung cancer from benign disease in pulmonary nodules, especially lesions with ground‐glass opacity (GGO), we assessed gene mutations of ctDNA in peripheral blood using targeted next‐generation sequencing (NGS).MethodsSingle pulmonary nodule patients without mediastinal lymph nodes and symptoms that were hard to diagnose by chest CT and lung cancer biomarker measurement in multiple medical centers were enrolled into the study. All patients accepted minimally invasive surgery but refused preoperative biopsy. Gene mutations in preoperative blood samples were detected by targeted NGS. Mutations with significant differences between lung tumors and benign lesions, as grouped by postoperative pathology, were screened. Protein expression was determined by immunohistochemistry. Highly expressed genes were selected as biomarkers to verify the mutations in peripheral blood.ResultsIn the training set, the RNF213, KMT2D, CSMD3 and LRP1B genes were mutated more frequently in early‐stage lung cancer (27 cases) than in benign nodules (15 cases) (P < 0.05). High expression of the RNF213 gene in lung cancers and low expression in benign diseases were seen by immunohistochemistry. The RNF213 gene was mutated in 25% of lung cancer samples in the validation set of 28 samples and showed high specificity (100%). In GGO patients, RNF213 was mutated more frequently in early‐stage lung cancer compared to benign diseases (P < 0.05).Conclusions RNF213 gene mutations were observed more frequently in early‐stage lung cancer, but not in benign nodules. Mutation of the RNF213 gene in peripheral blood may be a high specificity biomarker for the assisted early diagnosis of lung cancer in pulmonary nodules.Key pointsSignificant findings of the study: In peripheral venous blood and tumor tissue, RNF213 gene mutated more frequently in lung cancer than benign pulmonary nodules.What this study adds: Detection mutation of the RNF213 gene in peripheral blood may be a high specificity method for the assisted early diagnosis of lung cancer in pulmonary nodules.

The early detection, early diagnosis, and early treatment of lung cancer are the best strategies to improve the 5-year survival rate. Logistic regression analysis can be a helpful tool in the early detection of high-risk groups of lung cancer. Convolutional neural network (CNN) could distinguish benign from malignant pulmonary nodules, which is critical for early precise diagnosis and treatment. Here, we developed a risk assessment model of lung cancer and a high-precision classification diagnostic model using these technologies so as to provide a basis for early screening of lung cancer and for intelligent differential diagnosis. A total of 355 lung cancer patients, 444 patients with benign lung disease and 472 healthy people from The First Affiliated Hospital of Zhengzhou University were included in this study. Moreover, the dataset of 607 lung computed tomography images was collected from the above patients. The logistic regression method was employed to screen the high-risk groups of lung cancer, and the CNN model was designed to classify pulmonary nodules into benign or malignant nodules. The area under the curve of the lung cancer risk assessment model in the training set and the testing set were 0.823 and 0.710, respectively. After finely optimizing the settings of the CNN, the area under the curve could reach 0.984. This performance demonstrated that the lung cancer risk assessment model could be used to screen for high-risk individuals with lung cancer and the CNN framework was suitable for the differential diagnosis of pulmonary nodules.

Currently, digital healthcare technologies based on Internet of Things (IOT) are emerging, and the market for digital healthcare was estimated to be 151.5 billion dollars by 2020. Among the digital healthcare technologies, disease diagnosis is attracting maximum attention. Traditional diagnostic methods, such as radiology and human biological material analysis, have limitations of long study time, high cost, radiation exposure, and pain. Therefore, non-invasive diagnostics methods, for early diagnosis of specific diseases and personal healthcare management, have attracted attention.Breath sensors are one of the promising candidates for the realization of non-invasive diagnosis owing to their ease of use, safety, and cost effectiveness. The exhaled breath analysis technique holds great promise as a non-invasive method for diagnosing lung cancer. In the breath of lung cancer patients, various volatile organic compounds (VOCs) markers are generated during metabolic processes. Analyzing these VOCs markers presents a compelling avenue for early and non-invasive lung cancer diagnosis. Solid state gas sensors capable of measuring specific VOCs are currently being researched and utilized as commercial gas sensors for this purpose. However, the concentration of lung cancer VOCs markers in exhaled breath is commonly very low, and individual commercial gas sensors often lack the necessary sensitivity and selectivity to accurately diagnose lung cancer. In this study, we propose a multimodal solid state electronic biosensor system designed to differentiate lung cancer VOCs characteristics among multiple VOCs, enabling exhalation-based lung cancer diagnosis. Initially, we developed multimodal biosensors using more than 20 gas sensors, consisting of metal oxide semiconductor sensors, electrochemical sensors, photoionized detectors, metalorganic frame sensors and so on. And each exhibiting varying sensitivity profiles for complex gas mixtures. Additionally, we designed a classification algorithm based on convolutional neural networks (CNN) and multi-layered perceptron (MLP) to distinguish between lung cancer patients and healthy individuals using the output from these diverse multimodal gas sensors. To evaluate the system's performance, we collected exhaled air samples from 74 lung cancer patients and 107 normal subjects, conducting clinical trials with our developed system. We did model evaluation with confusion matrix and ROC curve. The results of these tests confirmed that our system achieved a high predictive accuracy of over 95% in classifying lung cancer patients. The simplicity of the sensor configuration in our developed system makes it suitable for large-scale screening tests to identify potential candidates for precise lung cancer diagnosis. This technology has the potential to contribute to reduced national welfare costs and improved patient survival rates through early diagnosis.

Lungs are the most crucial organs in a human body. Since the cancer detection began, lung cancer has been the most common terminal disease amongst all type of cancers. The contribution of deep learning, especially the convolution neural networks has widely reduced the mortality rates resulting from lung cancer. The classification of Computed Tomography (CT) images has enhanced the early diagnosis of lung cancer that has enabled victims to undergo treatment at an early stage. The resolution of the CT images have been variedly used for the accuracy of the model. Besides, the detection of lumps or anomalies in the images has greatly supported early diagnosis. Classification plays a vital role in the deep learning models to sort out the input images as positive and negative based on the attribute of the model built. However, the generalisation of classifiers has reduced the accuracy of the corresponding models built. To increase the accuracy and efficiency of the deep learning model, an optimised classification technique is used to predict lung cancer from the CT images. The purpose of optimisation here will enable the model to adapt stipulated feature extraction process according to the input images fed into the network. The model will be trained for predicting purpose given any resolution of the images. KEYWORDS: Lung cancer, CT images, Classification techniques, Optimised Classification, Prediction

Microsatellite alterations in bronchial and sputum specimens of lung cancer patients: Miozzo M, Sozzi G, Musso K et al. Division of Experimental Oncology A, Istituto Nazionale Tumori, Via G, Venezian 1, 20133 Milano. Cancer Res 1996;56:2285–2288

Early diagnosis of lung cancer is very important in improving patients life expectancies. Due to the high number of Computed Tomography (CT) images, fast and accurate diagnosis is difficult for radiologists. Therefore, there is an increasing demand for Computer-Aid Diagnosis (CAD) lung cancer. The core of all lung cancer detection systems is the distinction between cancer and non-cancerous tissues. This operation is performed in the false positive reduction phase, which is one of the most critical part of the lung cancer detection systems. The primary objective of this paper is to present a new method based on 3D Convolutional Neural Networks (CNN) that can reduce the false positives rate while providing a high sensitivity in detecting lung cancer lesions. We obtained 91.23% accuracy for 3.99 false positive per scan using a new method for fusion. The reason for accuracy improvement while reducing the false positive rate is by taking advantage of knowledge obtained from the classifiers in using a new fusion method.

Lung cancer remains one of the deadliest cancers worldwide, with survival rates significantly dependent on early detection. Traditional diagnostic methods, while effective, often identify lung cancer in advanced stages, limiting treatment options. Recently, artificial intelligence (AI) and deep learning (DL) models have gained attention for their potential to assist in early lung cancer diagnosis through radiological data, particularly computed tomography (CT) and X-ray imaging. This review examines recent advancements in AI and DL models specifically applied to early lung cancer detection, offering a comprehensive look at model architecture, data preprocessing techniques, and performance metrics. AI models, particularly those using convolutional neural networks (CNNs) and recurrent neural networks (RNNs), have shown significant promise in improving diagnostic accuracy. In studies published over the past two years, CNN-based models have achieved sensitivity rates upwards of 90% when applied to large datasets, such as the LIDC-IDRI, a publicly available database containing thousands of annotated lung scans. Furthermore, hybrid approaches combining AI models with radiologist expertise have demonstrated reduced false positives, a frequent challenge in automated image analysis. A critical factor in these advancements has been data quality and volume. Large, annotated datasets with high-resolution images have enabled more robust training and validation, essential for refining these models. However, challenges remain, particularly regarding the standardization of data across institutions, variations in scanner quality, and ethical concerns surrounding patient privacy. The review also highlights studies exploring multi-modal approaches, integrating radiological data with clinical records and genetic markers to create more personalized diagnostic tools. These multi-modal methods have the potential to improve predictive accuracy further, though they currently require more extensive validation. Overall, while AI and DL technologies offer transformative potential for early lung cancer detection, widespread implementation depends on continued research in model accuracy, data standardization, and ethical safeguards. This paper concludes by emphasizing the need for collaboration across disciplines—AI researchers, radiologists, and oncologists—to refine these models into reliable tools that can be integrated seamlessly into clinical workflows for timely and accurate lung cancer diagnosis.

CCSA-RF enhanced lightweight multi-scale CNN framework for robust lung cancer classification

Lung cancer is the leading cause of cancer-related death worldwide. The early detection of lung cancer is crucial for the diagnosis of this disease. Therefore, an effective and noninvasive method for the early diagnosis of lung cancer is urgently needed. To evaluate the diagnostic performance of circulating genetically abnormal cells (CACs) in early lung cancer, a total of 63 participants who completed CAC detection by Zhuhai SanMed Biotech Inc. and obtained pathological results from January to December 2020 were included in our study; 50 patients had lung cancer and 13 patients had benign lung disease. The levels of lung cancer-related markers in peripheral blood and the chest computed tomography (CT) imaging characteristics of these patients were collected before pathological acquisition. The positive rate of CAC was 90.0% in the lung cancer group and 23.1% in the benign lung disease group, and the difference was statistically significant (P < 0.01). The area under the receiver operating characteristic (ROC) curve of CAC was 0.837, the sensitivity was 90%, and the specificity was 76.9%. The area under the ROC curve and sensitivity were both higher than those of the combined or single serum tumor marker test. This study preliminarily concludes that the CAC test, as a noninvasive test, has high sensitivity and specificity for the early diagnosis of lung cancer. This test is expected to help with the early detection of disease in lung cancer patients.

Multi-center study on the application potential of Siaα-2,6Gal in early and differential diagnosis of lung cancer.

Lung cancer is one of the most threatening malignancies among the different kinds of tumors. The incidence and mortality rate are increasing especially in male. Advances in diagnosis and treatment have been achieve in recent years. However, the lung tumor cells also developing chemo- and radio-resistance. Novel approaches and new treatments are stilled needed to develop for early diagnosis and treatment. Recently, long non-coding RNAs (lncRNAs) original exosomes were proved different expression in lung tumor, which mediate multiple biological processes and is responsible for tumor proliferation and metastasis. In this review, we focus on the emerging roles of both lncRNAs and exosomal lncRNAs in lung cancer and their roles on angiogenesis, metastasis, diagnosis, drug resistance, and immune regulation of lung cancer. Exosome lncRNAs were proved to serve as regulatory factors for gene expression, mediating intercellular communication, and participating in the occurrence and development of various diseases. In addition, exosomes lnc RNA has advantages on the early diagnosis of lung cancer, tumor cell metastasis, drug resistance, and immune regulation. Exosome lncRNAs an provide some unique ideas on how to improve the efficiency of diagnosis and treatment of lung cancer in the future.

BackgroundMeasurement of tumor markers from peripheral venous blood is an emerging tool to assist in the early diagnosis of lung cancer. Samples from the pulmonary artery and pulmonary artery wedge position (trans-pulmonary samples) are accessible via right-heart catheterization and, by virtue of their proximity to lung tumors, may increase diagnostic yield.Case presentationWe report a case of a 64 year-old woman from whom trans-pulmonary samples were obtained and who was diagnosed 16 months later with recurrent metastatic small cell lung cancer. Carcinoembryonic antigen, cytokeratin fragment 21 − 1 (CYFRA), and human epididymis protein 4 (HE4) levels demonstrated increasing concentrations across the pulmonary circulation. These gradients exceeded the assays’ coefficient of variation by several-fold. For CYFRA and HE4, pulmonary artery wedge concentrations exceeded peripheral venous levels by more than 10% and peripheral arterial levels were up to 8% higher than peripheral venous levels.ConclusionsEvaluating the feasibility and utility of trans-pulmonary tumor markers for lung cancer diagnosis in a larger cohort should be considered. The addition of a peripheral arterial sample to standard peripheral venous samples may be a more practical alternative.

Advanced vapour sensing materials: Existing and latent to acoustic wave sensors for VOCs detection as the potential exhaled breath biomarkers for lung cancer

Lung cancer is a malicious tumor that originates in the respiratory tissues when the cells lining the airways of the lungs proliferate and expand uncontrollably. While computed tomography (CT) scans are capable of illuminating problematic areas, they are not proficient at independently diagnosing lung tumors. The task of accurately determining the nodule distribution within the lung CT scans through automatic identification is challenging. As a consequence, the delineation of lung CT images facilitates the identification and classification of lung tumors. This study involved the acquisition of 1097 lung CT image datasets from the Iraq-Oncology Teaching Hospital/National Centre for Cancer Diseases (IQ-OTH/NCCD). In the initial stage of the work, lung CT images are segmented to find malignancies utilizing the U-Net and Attention Gate Residual U-Net (AGRes U-Net) algorithms. The Intersection over Union (IoU) and binary focal loss metrics are applied to evaluate the segmented lung images. AGRes U-Net achieves 97% greater accuracy than the standard U-Net architecture, as determined by performance evaluation. Subsequently, a YOLOv5 network is implemented to annotate the segmented lung CT images with lesions. Thus, the segmented lung CT outputs are labelled for tumors by the YOLOv5 network are fed into the VGG-19 architecture. With an accuracy of 94.8%, this VGG-19 framework distinguishes the lung tumors as normal, benign, and malignant. To compare the performance of the segmented lung output to that of the classified output by means of a convolutional neural network (CNN), the labelled lung CT images are fed as input to a CNN model in the second phase of the work. In conjunction with the Adaptive Moment Estimation (Adam) optimizer, the CNN model is implemented. The Adam optimizer demonstrates a 98% accuracy rate for the classified tumor outputs. The study illustrates that the accuracy of AGResU-Net segmentation for lung cancer detection and classification, at 97%, is nearly equivalent to that of the CNN model, which achieves 98% accuracy.

Lung cancer is the leading cause of cancer death worldwide and the early diagnosis of lung cancer has recently become more important. For early screening lung cancer, computed tomography (CT) has been used as a gold standard for early diagnosis of lung cancer [1]. The major advantage of CT is that it is not susceptible to the problem of misdiagnosis caused by anatomical overlapping while CT has extremely high radiation dose and cost compared to chest radiography. Chest digital tomosynthesis (CDT) is a recently introduced new modality for lung cancer screening with relatively low radiation dose compared to CT [2] and also showing high sensitivity and specificity to prevent anatomical overlapping occurred in chest radiography. Dual energy material decomposition method has been proposed for better detection of pulmonary nodules as means of reducing the anatomical noise [3]. In this study, possibility of material decomposition in CDT was tested by simulation study and actual experiment using prototype CDT. Furthermore organ absorbed dose and effective dose were compared with single energy CDT. The Gate v6 (Geant4 application for tomographic emission), and TASMIP (Tungsten anode spectral model using the interpolating polynomial) code were used for simulation study and simulated cylinder shape phantom consisted of 4 inner beads which were filled with spine, rib, muscle and lung equivalent materials. The patient dose was estimated by PCXMC 1.5 Monte Carlo simulation tool [4]. The tomosynthesis scan was performed with a linear movement and 21 projection images were obtained over 30 degree of angular range with 1.5° degree of angular interval. The proto type CDT system has same geometry with simulation study and composed of E7869X (Toshiba, Japan) x-ray tube and FDX3543RPW (Toshiba, Japan) detector. The result images showed that reconstructed with dual energy clearly visualize lung filed by removing unnecessary bony structure. Furthermore, dual energy CDT could enhance spine bone hidden by heart effectively. The effective dose in dual energy CDT was slightly higher than single energy CDT, while only 10% of average thoracic CT [5]. Dual energy tomosynthesis is a new technique; therefore, there is little guidance for its integration into the clinical practice and this study can be used to improve diagnosis efficiency of lung field screening using CDT