Lung cancer classification using data mining and supervised learning algorithms on multi-dimensional data set

Comments: This publication provides long-term followup in a group of 32 patients who underwent bronchoscopic treatment (BT) for the treatment of intraluminal microinvasive radiographically occult lung cancer (ROLC). These patients were not eligible for surgical resection because of comorbidities and other reasons. These 32 patients (28 men and 4 women) met the following inclusion criteria for the study: surgically nonresectable candidates because of poor lung function or the presence of multiple primaries; a strictly ≤ 1 cm in size, intraluminal, microinvasive ROLC (carcinoma in situ excluded) with visible distal tumor margin by conventional and autofluorescence bronchoscopy; no bronchial wall invasion or extraluminal tumor growth on highresolution computed tomography. History of prior primary lung cancer was present in 13 patients, and these were surgically resected as long as 13 years before current synchronous or metachronous primaries. The TNM stages of previous resections for the non-ROLC tumors were stage IB in seven patients, stage IIB in four patients, stage IIIA in one patient, and stage IIIB in one patient. There was no sign of recurrence or metastatic disease of the previous non-ROLC cancer before BT. In six patients, multiple ROLCs were found at different sites in both lungs (four patients with two lesions and two patients with three lesions) that met all criteria for BT as described earlier. BT was administered with intent to cure cancer, and consecutive patients were treated with photodynamic therapy (n = 5), Nd:YAG laser therapy (n = 1), electrocautery (n = 24), and argon plasma coagulation (n = 2). Follow-up evaluation at 3 to 4 months interval included high-resolution computed tomographic scans, both conventional and autofluorescence bronchoscopy, including biopsy specimens and brush cytology for histologic evaluation. The average follow-up period was 5 years (range, 2–10 years). In three patients local recurrence was again successfully treated with electrocautery. Sixteen patients (50%) died during follow-up. Eight of the nine patients who died as a result of lung cancer had a previous resection of a more advanced stage lung cancer as long as 5 years before BT of the ROLC. In the remaining seven patients, the cause of death was not related to lung cancer. Sixteen patients are still alive without any tumor recurrence. Based on these observations, the authors conclude that BT is an effective treatment modality for high-risk patients with ROLC who are not eligible for surgical resection. Early detection of lung cancer by currently available methods such as low-dose spiral computed tomography, screening sputum cytology, autofluorescence bronchoscopy, and other techniques may enable surgical cure in a number of patients. However, not all patients are surgical candidates because of various reasons. One of the main reasons is the presence of cancer in the major airways (trachea and within 2.0 cm of the main carina), which upgrades the cancer stage. More commonly, surgery and anesthesia are contraindicated because of underlying medical conditions such as severe chronic obstructive pulmonary disease, severe malnutrition, and other conditions. When such patients develop intraluminal microinvasive ROLC in the central part of the tracheabronchial tree, various types of BT described in this publication may benefit them. Nonetheless, as noted by the authors of this study, early detection of lung cancer and BT in such patients is questionable because this expensive and time-consuming exercise may have little or no impact on mortality. Death from lung cancer occurred in 9 patients (28%), and death from other causes was noted in 7 patients (22%). The positive outcome of this study is that 50% of patients were still alive after BT, with an average follow-up of 5 years (range, 2–10 years).

Exosomal lipids for classifying early and late stage non-small cell lung cancer

A review of detection methods for the early diagnosis of lung cancer

The examination and treatment of lung cancer in its early stages rely heavily on diagnostic imaging technologies. Various imaging procedures used in healthcare, such as computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), chest x-ray, and molecular imaging (MI) approaches, have been extensively studied for the detection and classification of lung cancer. However, some of these systems have limitations, such as their inability to accurately classify lung cancer images, rendering them unsuitable for use in individuals with other illnesses. There is a critical need for the development of a robust and precise method for the early detection of lung cancer. One of the fastest-growing areas in health imaging is deep learning (DL), which involves rapidly developing techniques that utilize texture and clinical-image-based information. DL-based clinical imaging technologies may assist physicians in recognizing and classifying lung cancers more easily and accurately. In this paper, a deep ensemble model is proposed for detecting and classifying lung cancer based on three DL classifiers: MobileNet, ResNet50, and XceptionNet. These DL models are trained on a publicly available lung cancer dataset containing normal, benign, and malignant images. The performance of these baseline classifiers can be measured in terms of accuracy, precision, and recall. The proposed ensemble model accurately detects and classifies lung cancer images. The final results show classification accuracy scores of 28.00%, 13.00%, and 36.00% for the baseline classifiers MobileNet, ResNet50, and XceptionNet, respectively, over the dataset. The proposed ensemble model outperforms the baseline classifiers, achieving a classification accuracy score of 99.00%. Finally, the performance of the proposed ensemble model is compared with those of existing state-of-the-art methods.

8553 Background: Lung cancer early detection improves survival, but risk-based low-dose CT screening (LDCT) only identifies a minority of patients. We implemented an ILNP in a community healthcare system, and evaluated its risks and benefits. Methods: Patients with lung lesions on routinely-performed radiologic studies were flagged by radiologists and triaged using evidence-based guidelines. We tracked demographics, clinical characteristics, procedures, complications, and health outcomes. We analyzed ILNP subjects’ eligibility for LDCT by National Lung Screening Trial (NLST), Center for Medicaid Services (CMS), NEderlands Leuvens Screening ONderzoek (NELSON), National Comprehensive Cancer Network (NCCN) Risk Groups 1 and 2 (screening recommended), NCCN Risk Group 3 (screening not currently recommended), and US Preventive Services Task Force (USPSTF) criteria from 2013 and 2020. Statistical analysis used the chi-square test and Kaplan Meier method. Results: From 2015-2020, 13,710 patients were evaluated in the ILNP program: median age, 64 years; 42% male; 65% White, 29% Black; 667 (4.9%) were diagnosed with lung cancer. Lung cancers diagnosed from ILNP were 39% adenocarcinoma / 20% Squamous Cell with clinical stage distribution 49% I, 8% II, 17% III, and 16% IV. 832 (6.1%) had invasive diagnostic testing- CT-guided biopsy (50%), bronchoscopy (30%), and/or EBUS (26%); 11% of the 832 had >1 invasive diagnostic test. The most common complications from invasive testing were pneumothorax and chest tube placement. Only 11%-20% of all ILNP patients would have been eligible for LDCT. In ILNP patients diagnosed with lung cancer, only 33% were eligible for screening by NLST criteria; the proportion increased substantially when USPSTF 2020 or NCCN Group 2 criteria were applied (Table). Compared to NLST, NCCN Group 2 criteria increased screening eligibility among cancer patients by 22% (from 33% to 55%), while only increasing screening eligibility by 6% (from 8% to 14%) in non-cancer patients. Aggregate 1-year and 3-year survival rates for lung cancer patient diagnosed through ILNP were 76% (95% CI: 73, 80) and 64% (95% CI: 59, 69). Conclusions: The ILNP identified early-stage lung cancer more frequently than most LDCT programs, with promising survival rates. The majority of subjects with lung cancer were not eligible for LDCT, we still need to optimize risk-based screening criteria. Even with new, expanded criteria for LDCT, structured ILNP is necessary to expand early detection of lung cancer.[Table: see text]

Early detection of lung cancer remains a critical challenge in medical imaging, where accurate classification between benign and malignant pulmonary nodules can significantly impact patient outcomes. This study presents a comprehensive comparative analysis of six distinct machine learning approaches for lung cancer classification using 3D computed tomography (CT) data. The methodologies evaluated include 2D convolutional neural networks (CNN), 3D CNN architectures, multi-slice processing techniques, and traditional machine learning algorithms including Random Forest and Logistic Regression. Our experimental framework utilized synthetic CT volumes designed to replicate characteristic patterns of benign and malignant lung lesions. The synthetic dataset comprised 120 volumetric samples with distinct spatial patterns representing clinical features such as nodule morphology, tissue density variations, and anatomical distribution. Results demonstrate that traditional machine learning approaches, particularly Random Forest classifiers, achieved superior performance with accuracy rates exceeding 85%, while deep learning models showed variable performance depending on architecture complexity and data preprocessing strategies. The findings suggest that careful consideration of model architecture selection and data representation is crucial for effective lung cancer classification systems.

Early detection of lung cancer is crucial for patient survival and treatment. Recent advancements in next-generation sequencing (NGS) analysis enable cell-free DNA (cfDNA) liquid biopsy to detect changes, like chromosomal rearrangements, somatic mutations, and copy number variations (CNVs), in cancer. Machine learning (ML) analysis using cancer markers is a highly promising tool for identifying patterns and anomalies in cancers, making the development of ML-based analysis methods essential. We collected blood samples from 92 lung cancer patients and 80 healthy individuals to analyze the distinction between them. The detection of lung cancer markers Cyfra21 and carcinoembryonic antigen (CEA) in blood revealed significant differences between patients and controls. We performed machine learning analysis to obtain AUC values via Adaptive Boosting (AdaBoost), Multi-Layer Perceptron (MLP), and Logistic Regression (LR) using cancer markers, cfDNA concentrations, and CNV screening. Furthermore, combining the analysis of all multi-omics data for ML showed higher AUC values compared with analyzing each element separately, suggesting the potential for a highly accurate diagnosis of cancer. Overall, our results from ML analysis using multi-omics data obtained from blood demonstrate a remarkable ability of the model to distinguish between lung cancer and healthy individuals, highlighting the potential for a diagnostic model against lung cancer.

Lung cancer remains one of the leading causes of death worldwide, with many sufferers unaware of their condition until it is too late for treatment. Therefore, high-accuracy prediction methods are urgently needed for early detection of lung cancer. This research uses the Random Forest algorithm, known for its excellent performance in medical data classification. In this study, modeling was optimized by implementing hyperparameter optimization using Optuna. The results of the generated model show an accuracy rate of 98.6%, which is highly significant in the context of early lung cancer detection. Additionally, this algorithm demonstrated 100% recall for the positive class and 97% for the negative class, indicating that the model is highly effective in identifying patients who truly have lung cancer. Another advantage of this model is seen in the AUC (Area Under the Curve) value reaching 1, indicating 100% accurate predictions. With these results, this research affirms the importance of using the Random Forest algorithm in developing early detection systems for lung cancer. This not only can improve treatment success rates but also significantly reduce mortality rates from lung cancer.

Accurate classification and early detection of lung cancer are crucial for effective treatment and improved patient outcomes. Although recent advances in medical imaging have improved diagnostic workflows, many traditional approaches remain limited in their ability to efficiently analyze large volumes of imaging data. Accordingly, this study proposes a deep learning framework for lung cancer classification using Digital Imaging and Communications in Medicine (DICOM) images. The proposed approach integrates an Intelligent Squirrel Search (ISS)–tuned VGG16 network (ISS-VGG16) to improve classification performance. A set of lung cancer DICOM images was obtained from an open-source dataset. Preprocessing steps, including image resizing and contrast enhancement, were applied to standardize the inputs for model training. ISS was used to optimize key hyperparameters of the VGG16 model, thereby improving classification performance. The model was implemented in Python. Experimental results indicate that the proposed ISS-VGG16 approach outperforms baseline methods, achieving a recall of 0.92, precision of 0.96, F1-score of 0.95, and overall accuracy of 0.97 on the evaluated lung cancer image dataset. These results demonstrate reliable classification performance across the defined lung cancer classes. Overall, the proposed framework provides an effective and dependable approach for lung cancer image classification.

Lung cancer mortality and morbidity rates are the first among malignant tumors. It is extremely crucial to pay more attention to the early diagnosis and treatment of lung cancer and to grasp and judge the progress of the patient's condition promptly. In this study, lung cancer patients' early diagnosis with tumor markers and inflammatory variables is examined. The general surgery department of our hospital treated 98 patients with lung cancer and 100 patients with benign pulmonary hyperplasia from January 2017 to February 2018. Additionally, 100 healthy subjects who completed physical examinations during this time period were included. Based on the findings of the pathological examination, 100 patients with benign pulmonary hyperplasia were chosen as the benign group, 100 patients with lung cancer were chosen as the malignant group, and 100 healthy individuals were chosen as the healthy group. The tumor markers carbohydrate antigen 125 (CA125), CA153, and carcinoembryonic antigen (CEA), as well as inflammatory factors such as tumor necrosis factor- (TNF-) and high-sensitivity C-reactive protein, were measured in the venous blood of three groups of patients (hs-CRP). There was no discernible difference in tumor marker levels between the benign and healthy groups (P > 0.05). In comparison to the benign and healthy groups, the malignant group had higher serum levels of CA153, CA125, and CEA (P 0.05). Between the benign and healthy groups, there was no discernible difference in the levels of inflammatory factors (P > 0.05). TNF- and hs-CRP serum levels were observed to be higher in the malignant group compared to the benign and healthy groups (P 0.05). The combined detection of CA153 + CA125 + CEA + TNF − +hs − CRP showed the highest sensitivity and specificity, which were, respectively, 62.22 percent and 92.00 percent, when compared to single or mixed detection of tumor markers or inflammatory factors solely. Serum levels of inflammatory agents TNF- and hs-CRP may be related to the pathophysiology and other functions of patients with lung cancer, as well as to the development and metastasis of the disease. These markers include CA153, CA125, and CEA. For the early detection of lung cancer and the evaluation of the disease's severity, the detection of tumor markers in combination with inflammatory variables has a significant reference value.

In this meta-analysis, we evaluated the usefulness of serum thymidine kinase 1 concentration (STK1c) for monitoring the outcome of extensive open surgery in patients with lung cancer. We also compared STK1c between a healthy population and patients with benign and malignant lung tumors to assess its potential value for early detection of lung cancer and for distinguishing between benign lung disease and malignant cancer. Related studies were retrieved from publications in PubMed, Cochrane, China National Knowledge Infrastructure, Wanfang databases, and Internet searches. Correlation was evaluated using weighted mean difference. Fixed or random effect models were selected for data analyses based on heterogeneity tested with the chi-square test. Publication bias was assessed using a funnel plot and Egger's test. Twenty studies were selected for analysis, which showed that STK1c was significantly (p < 0.00001) reduced by 41.7% 1 month after extensive open surgery, approximately corresponding to an STK1c half-life of 1 month. STK1c levels were significantly higher in lung cancer patients than in healthy persons (p < 0.00001) or in patients with benign lung disease (p < 0.00001). There was also a significant difference in STK1c between patients with benign and malignant lung disease (p < 0.0001). The half-life of STK1c may be an important tool in the clinical evaluation of surgical response in patients with lung cancer. STK1c may also be beneficial in the early detection of lung cancer.

Lung cancer is the leading cause of cancer death worldwide. Survival is largely dependent on the stage of diagnosis: the localized disease has a 5-year survival greater than 55%, whereas, for spread tumors, this rate is only 4%. Therefore, the early detection of lung cancer is key for improving prognosis. In this study, we present an innovative, non-invasive, cancer detection approach based on measurements of the metabolic activity profiles of immune system cells. For each Liquid ImmunoBiopsy test, a 384 multi-well plate is loaded with freshly separated PBMCs, and each well contains 1 of the 16 selected stimulants in several increasing concentrations. The extracellular acidity is measured in both air-open and hermetically-sealed states, using a commercial fluorescence plate reader, for approximately 1.5h. Both states enable the measurement of real-time accumulation of 'soluble' versus 'volatile' metabolic products, thereby differentiating between oxidative phosphorylation and aerobic glycolysis. The metabolic activity profiles are analyzed for cancer diagnosis by machine-learning tools. We present a diagnostic accuracy study, using a multivariable prediction model to differentiate between lung cancer and control blood samples. The model was developed and tested using a cohort of 200 subjects (100 lung cancer and 100 controlsubjects), yielding 91% sensitivity and 80% specificity in a 20-fold cross-validation. Our results clearly indicate that the proposed clinical model is suitable for non-invasive early lung cancer diagnosis, and is indifferent to lung cancer stage and histological type.

Background: Our feasibility study employed a novel genomic detection methodology that enriches 5-hydroxymethylcytosine (5hmC) loci in cell free DNA (cfDNA) from the plasma of cancer patients using click chemistry coupled with sequencing and machine learning based classification methods. These classification methods were developed to detect the presence of disease in the plasma of cancer and control subjects. Cancer and control patient cfDNA cohorts were accrued from multiple sites consisting of 48 breast, 55 lung, 32 prostate and 2 pancreatic datasets consisting of 41 and 53 cancer subjects (Set 1 and 2). In addition, a control cohort of 260 subjects (non-cancer) was employed to match cancer patient demographics (age, sex and smoking status) in a case-control study design. Methods: Machine learning methods, applied to each cancer case cohort individually, with a balancing non-cancer cohort, were able to classify cancer and control samples. Measures of predictive performance using 5-fold cross validation coupled with out-of-fold Area Under the Receiver Operating Characteristic Curve (AUROC) measures were employed. Gene sets selected as part of biomarker discovery were further analyzed for disease relevance using pathway analysis tools (GSEA, mSigDB). Results: 260 controls and 229 cancers from four disease types (breast, lung, pancreas and prostate) were analyzed; more than 60% of cancer patients had early stage disease (I or II). Predictive performance, employing AUROC measures, was established for breast (0.89), lung (0.84), pancreas (set 1 - 0.95 and 2 - 0.93) and prostate (0.83). The genes defining each of these predictive models were enriched for pathways relevant to disease specific etiology, notably in the control of gene regulation in these same pathways. The breast cancer cohort consisted primarily of stage I and II patients including tumors &amp;lt; 2 cm and these samples exhibited a higher prediction probability score. The prostate cancer cohort consisted of both indolent and aggressive disease sample and prediction performance was equally high for both (AUROC for indolent vs aggressive was 0.81 and 0.77, respectively). Conclusions: These findings suggest that 5hmC changes in cfDNA enable non-invasive detection of early stage breast, pancreatic, prostate, and lung cancers. Furthermore, 5hmC profiling in cfDNA may enable the prediction of clinically relevant features such as tumor size in breast adenocarcinoma or indolent disease in prostate cancer. Finally, this study identified a suite of 5hmC biomarkers that may be further validated in larger, and more diverse, patient cohorts. Citation Format: Anna Bergamaschi, Jeremy Ku, Yuhong Ning, Francois Collin, Chris Ellison, Tierney Phillips, Erin McCarthy, Wendy Wang, Michael Antoine, David Haan, Aaron Scott, Paul Lloyd, Gulfem Guler, Alan Ashworth, Stephen Quake, Samuel Levy. Epigenomic detection of multiple cancers in plasma derived cell free DNA [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 783.

Lung cancer kills millions of people each year in the world and is the main cause of cancer death in men and the second cause in women (1). Surprisingly, a decade ago, lung cancer was not so frequently diagnosed compared to breast and prostate cancer, explained by the fact that these cancers can be detected at a very early stage. For lung cancer, the five-year survival rate was as low as 15%. However, when lung cancer can be detected at an early stage, the survival rate increases dramatically (2). Using low-dose computed tomography (CT) screening, a 20% reduction in lung cancer mortality has been more recently reported (3). Thus, during this last decade, many efforts have been made to develop innovative technologies to detect lung cancer very early, with a special focus on molecular markers as an alternative approach to different imaging and cytology-based strategies.

Lung cancer is a leading cause of death in Chile, where late-stage diagnoses and high mortality rates prevail. Here, we describe the development of OncovigIA, a novel digital tool powered by natural language processing that enhances the identification of potential lung cancer cases by surveilling computed tomography (CT) reports in a large public Hospital in Santiago, Chile. We combined natural language processing and large language models with state-of-the-art machine learning techniques and approaches to treat unbalanced data sets and determine the best solution to implement in OncovigIA. Focusing on key sections of the reports and using various machine learning models, including a balanced Random Forest, the tool achieved high performance with 0.90 accuracy and 0.84 F1-score on the test set. When applied to 13,326 CT chest reports from 2022, it successfully identified 377 CTs of patients with suspected lung cancer previously undetected and not managed by the multidisciplinary local lung cancer team. This study underscores the potential of artificial intelligence in early cancer detection and highlights the importance of its integration into local health care ecosystems. By promptly increasing the number of patients referred for specialized management, the tool OncovigIA offers a promising path toward improving lung cancer survival rates in Chile and beyond. Moreover, this article provides avenues for its broader implementation, extending it to other cancer types and/or health care-related texts for continuous surveillance, aiming at the early referral and treatment of cancer in low-resource settings.