Artificial intelligence-based prediction of clinical outcome in immunotherapy and targeted therapy of lung cancer

Seminars in Cancer Biology, Год журнала: 2022, Номер 86, С. 146 - 159

Опубликована: Авг. 11, 2022

Язык: Английский

---

CT-based delta-radiomics nomogram to predict pathological complete response after neoadjuvant chemoradiotherapy in esophageal squamous cell carcinoma patients

Journal of Translational Medicine, Год журнала: 2024, Номер 22(1)

Опубликована: Июнь 18, 2024

Abstract Background This study developed a nomogram model using CT-based delta-radiomics features and clinical factors to predict pathological complete response (pCR) in esophageal squamous cell carcinoma (ESCC) patients receiving neoadjuvant chemoradiotherapy (nCRT). Methods The retrospectively analyzed 232 ESCC who underwent pretreatment post-treatment CT scans. Patients were divided into training (n = 186) validation 46) sets through fivefold cross-validation. 837 radiomics extracted from regions of interest (ROIs) delineations on images before after nCRT calculate delta values. LASSO algorithm selected (DRF) based classification performance. Logistic regression constructed incorporating DRFs factors. Receiver operating characteristic (ROC) area under the curve (AUC) analyses evaluated performance for predicting pCR. Results No significant differences existed between datasets. 4-feature signature (DRS) demonstrated good predictive accuracy pCR, with α-binormal-based empirical AUCs 0.871 0.869. T-stage ( p 0.001) differentiation degree 0.018) independent predictors combined DRS improved dataset (AUC αbin 0.933 AUC emp 0.941). set showed similar 0.958 0.962. Conclusions provided high pCR nCRT.

Язык: Английский

---

Radiomics and Deep Learning Prediction of Immunotherapy-Induced Pneumonitis From Computed Tomography

JCO Clinical Cancer Informatics, Год журнала: 2025, Номер 9

Опубликована: Фев. 1, 2025

Primary barriers to application of immune checkpoint inhibitor (ICI) therapy for cancer include severe side effects (such as potentially life threatening pneumonitis [PN]), which can cause the discontinuation treatment. Predicting patients may develop PN while on ICI would improve both safety and potential efficacy because treatments could be safely administered longer or discontinued before toxicity. Starting from a cohort 3,351 with who received previous at Vanderbilt University Medical Center, we curated 2,700 contrast chest computed tomography (CT) volumes 671 patients. Three different pure imaging models predicted using only single time point first dose. The model used 109 radiomics features achieved an AUC 0.747 (CI, 0.705 0.789) positive predictive value (PPV) 0.244 0.211 0.276) sensitivity 0.553 0.485 0.621) mainly describing global lung properties. second convolutional neural network (CNN) raw CTs 0.819 0.781 0.857) PPV 0.203 0.284) 0.743 0.681 0.806). third combined deep learning but, 0.829 0.797 0.862) 0.254 0.228 0.281) 0.780 0.721 0.840), did not show significant improvement CNN-only model. This new suggests utility in prediction over traditional promises better management receiving ability stratify immunotherapy drug trials.

Язык: Английский

---

The influence of image selection and segmentation on the extraction of lung cancer imaging radiomics features using 3D-Slicer software

BMC Cancer, Год журнала: 2025, Номер 25(1)

Опубликована: Апрель 17, 2025

Extracting image features can predict the prognosis and treatment effect of non-small cell lung cancer, which has been increasingly confirmed. However, specific operation using 3D-Slicer still lacks standardization. For example, segmentation is manually performed based on window or automatically through mediastinal window. The images used for feature extraction are either enhanced plain scanned. It questionable whether these influencing factors will affect results be affected. This article intends to preliminarily explore above issues. downloaded 22 patients with cancer from Cancer Imaging Archive (TCIA), including 11 cases adenocarcinoma squamous carcinoma. Perform tumor scan image, image. Manual drawing window, automatic make manual modifications. radiomics Python radiomics. Firstly, analyze original sequence perform Shapiro test. If it follows a normal distribution, an analysis variance. does not follow Friedman Compare significantly different pairwise. Then, preliminary was conducted differences between carcinoma in each group. A total 88 sets imaging were extracted, 107 Among them, 33 showed significant differences. Continuing pairwise repeated testing, found that there 2 windows. There 12 windows one difference scanning enhancement 14 groups. scan. 13 According pathological grouping 54 adenocarcinoma. CT relatively small impact extracting features, while selecting features. Therefore, choosing should carefully considered, as size range also significant, indicating high possibility distinguishing (Liu C, He Y, Luo J, Influence Image Selection Segmentation Extraction Lung Radiomics Features Using Software, 2024).

Язык: Английский

---

Immune-related pulmonary toxicities of checkpoint inhibitors in non-small cell lung cancer: Diagnosis, mechanism, and treatment strategies

Frontiers in Immunology, Год журнала: 2023, Номер 14

Опубликована: Апрель 4, 2023

Immune checkpoint inhibitors (ICI) therapy based on programmed cell death-1 (PD-1) and death ligand 1 (PD-L1) has changed the treatment paradigm of advanced non-small lung cancer (NSCLC) improved survival expectancy patients. However, it also leads to immune-related adverse events (iRAEs), which result in multiple organ damage. Among them, most common one with highest mortality NSCLC patients treated ICI is inhibitor pneumonitis (CIP). The respiratory signs CIP are highly coincident overlap those primary cancer, causes difficulties detecting, diagnosing, managing, treating. In clinical management, serious should receive immunosuppressive even discontinue immunotherapy, impairs benefits ICIs potentially results tumor recrudesce. Therefore, accurate diagnosis, detailedly dissecting pathogenesis, developing reasonable strategies for essential prolong patient expand application ICI. Herein, we first summarized diagnosis NSCLC, including classical radiology examination rising serological test, pathology artificial intelligence aids. Then, dissected potential pathogenic mechanisms CIP, disordered T subsets, increase autoantibodies, cross-antigens reactivity, role other immune cells. Moreover, explored therapeutic approaches beyond first-line steroid future direction targeted signaling pathways. Finally, discussed current impediments, trends, challenges fighting ICI-related pneumonitis.

Язык: Английский

---

Enhanced CT-based radiomics model to predict natural killer cell infiltration and clinical prognosis in non-small cell lung cancer

Frontiers in Immunology, Год журнала: 2024, Номер 14

Опубликована: Янв. 12, 2024

Natural killer (NK) cells are crucial for tumor prognosis; however, their role in non-small-cell lung cancer (NSCLC) remains unclear. The current detection methods NSCLC inefficient and costly. Therefore, radiomics represent a promising alternative. We analyzed the radiogenomics datasets to extract clinical, radiological, transcriptome data. effect of NK on prognosis was assessed. Tumors were delineated using 3D Slicer, features extracted pyradiomics. A model developed validated five-fold cross-validation. nomogram constructed selected clinical variables radiomic score (RS). CIBERSORTx database gene set enrichment analysis used explore correlations cell infiltration molecular mechanisms. Higher correlated with better overall survival (OS) (P = 0.002). showed an area under curve 0.731, 0.726 post-validation. RS differed significantly between high low < 0.01). nomogram, variables, effectively predicted 3-year OS. ICOS BTLA genes 0.001) macrophage M0/M2 levels. key pathways included TNF-α signaling via NF-κB Wnt/β-catenin signaling. Our accurately NSCLC. Combined characteristics, it can predict patients Bioinformatic revealed expression underlying

Язык: Английский

---

CCL18, CHI3L1, ANG2, IL-6 systemic levels are associated with the extent of lung damage and radiomic features in SARS-CoV-2 infection

Inflammation Research, Год журнала: 2024, Номер 73(4), С. 515 - 530

Опубликована: Фев. 3, 2024

Язык: Английский

---

Deep learning for predicting the risk of immune checkpoint inhibitor-related pneumonitis in lung cancer

Clinical Radiology, Год журнала: 2023, Номер 78(5), С. e377 - e385

Опубликована: Янв. 14, 2023

Язык: Английский

---

CT Image-Based Radiomic Analysis for Detecting PD-L1 Expression Status in Bladder Cancer Patients

Academic Radiology, Год журнала: 2024, Номер unknown

Опубликована: Март 1, 2024

Rationale and Objectives

The role of Programmed death-ligand 1 (PD-L1) expression is crucial in guiding immunotherapy selection. This study aims to develop evaluate a radiomic model, leveraging Computed Tomography (CT) imaging, with the objective predicting PD-L1 status patients afflicted bladder cancer.

Materials Methods

encompassed 183 subjects diagnosed histologically confirmed cancer, among which PD-L1(+) cohort constituted 60.1% total population. Stratified random sampling was utilized at 7:3 ratio. We employed five diverse machine learning algorithms—Decision Tree, Random Forest, Linear Support Vector Classification, Machine, Logistic Regression—to establish models on training dataset. These endeavored predict premised features derived from region-of-interest segmentation. Subsequent this, predictive performance these examined validation set employing receiver operating characteristic (ROC) curve. DeLong test contrast ROC curves, thereby pinpointing model superior accuracy.

Results

16 were chosen for construction. All revealed strong (AUC, 0.920–1) commendable ability 0.753–0.766). As per test, no statistically significant disparities observed any (P > 0.05) set. Additional verification through calibration curve decision analysis indicated that Regression exhibited extraordinary precision practicality.

Conclusion

Our grounded features, demonstrated its proficiency accurately distinguishing cancer high expression. Future research, incorporating more exhaustive datasets, could potentially augment efficiency algorithms, advancing their clinical utility.

Язык: Английский

---

Application of Chest CT Imaging Feature Model in Distinguishing Squamous Cell Carcinoma and Adenocarcinoma of the Lung

Cancer Management and Research, Год журнала: 2024, Номер Volume 16, С. 547 - 557

Опубликована: Июнь 1, 2024

Purpose: In situations where pathological acquisition is difficult, there a lack of consensus on distinguishing between adenocarcinoma and squamous cell carcinoma from imaging images, each doctor can only make judgments based their own experience. This study aims to extract features chest CT, sensitive factors through logistic univariate multivariate analysis, model distinguish lung adenocarcinoma. Methods: We downloaded CT scans with clear diagnosis The Cancer Imaging Archive (TCIA), extracted 19 by radiologist thoracic surgeon, including location, spicule, lobulation, cavity, vacuolar sign, necrosis, pleural traction vascular bundle air bronchogram calcification, enhancement degree, distance pulmonary hilum, atelectasis, hilum bronchial lymph nodes, mediastinal interlobular septal thickening, metastasis, adjacent structures invasion, effusion. Firstly, we apply the glm function R language perform analysis all variables select P < 0.1. Then, selected obtain predictive model. Next, use roc in calculate AUC value draw ROC curve, val.prob Calibrat rmda package DCA curve clinical impact curve. At same time, 45 patients diagnosed surgery or biopsy Radiotherapy Department Thoracic Surgery our hospital 2023 2024 were included validation group. jointly determined recorded two doctors mentioned above image feature data are complete does not require preprocessing, so directly entering statistical calculations. Perform curves, calibration DCA, curves group further validate If performs well group, nomogram demonstrate. Results: 75 TCIA finally 18 for analysis. First, performed, total 5 obtained: Sign, nodes. After conducting modeling = 0.887, was established using cases hospital, Draw 0.865 evaluate accuracy Calibrate reliability practice practicality Conclusion: It possible influential ordinary determine carcinoma. have set up terms discrimination, accuracy, reliability, practicality. Keywords: cancer, LUAD, LSCC, features, predict

Язык: Английский

---