Non-invasive classification of non-small cell lung cancer: a comparison between random forest models utilising radiomic and semantic features.

Non-invasive classification of non-small cell lung cancer: a comparison between random forest models utilising radiomic and semantic features. Br J Radiol. 2019 Jun 05;:20190159 Authors: Bashir U, Kawa B, Siddique M, Mak SM, Nair A, Mclean E, Bille A, Goh V, Cook G Abstract OBJECTIVE: Non-invasive distinction between squamous cell carcinoma and adenocarcinoma subtypes of non-small-cell lung cancer (NSCLC) may be beneficial to patients unfit for invasive diagnostic procedures or when tissue is insufficient for diagnosis. The purpose of our study was to compare the performance of random forest algorithms utilizing CT radiomics and/or semantic features in classifying NSCLC. METHODS: Two thoracic radiologists scored 11 semantic features on CT scans of 106 patients with NSCLC. A set of 115 radiomics features was extracted from the CT scans. Random forest models were developed from semantic (RM-sem), radiomics (RM-rad), and all features combined (RM-all). External validation of models was performed using an independent test data set (n = 100) of CT scans. Model performance was measured with out-of-bag error and area under curve (AUC), and compared using receiver-operating characteristics curve analysis on the test data set. RESULTS: The median (interquartile-range) error rates of the models were: RF-sem 24.5 % (22.6 - 37.5 %), RF-rad 35.8 % (34.9 - 38.7 %), and RM-all 37.7 % (37.7 - 37.7). On training data, both RF-rad and RF-all gave perfect discrim...
Source: The British Journal of Radiology - Category: Radiology Authors: Tags: Br J Radiol Source Type: research

Related Links:

Conclusion: In view of the poor specificity, the current study calls for reconsideration of the universal recommendation of dedicated brain imaging (in addition to PET/CT scan) among NSCLC patients beyond stage I. PMID: 31116030 [PubMed - as supplied by publisher]
Source: Journal of Comparative Effectiveness Research - Category: General Medicine Tags: J Comp Eff Res Source Type: research
Conclusions: Hypermetabolic changes on DTP 18F-FDG PET showed good correlation with tumor positive expression of PD-L1 and high density of CD8 TILs and PD-1 in tumor immune microenvironment. Furthermore, dSUVmax was an independent predictor of PD-L1 expression in pretreated NSCLC.
Source: Journal of Nuclear Medicine - Category: Nuclear Medicine Authors: Tags: Lung Cancer I: Molecular Biomarkers and Molecular Imaging Signatures Source Type: research
Conclusions: Baseline MTV based parameters evaluated from primary tumor as well as the whole body tumor lesions are reliable prognostic markers of OS in NSCLC patients. Also, among the post-therapy parameters and percentage changes, whole body MTV based parameters i.e. post-TLGwb and TLGwb can predict OS of NSCLC patients.However, SUV based whole body PET/CT parameters (SUVmax, SUVavg, and TBR) evaluated from baseline as well as post-therapy scans have no prognostic value in these patients.
Source: Journal of Nuclear Medicine - Category: Nuclear Medicine Authors: Tags: Lung Cancer Posters Source Type: research
Conclusions: Locally stationary wavelet field texture analysis model can differentiate responders and non-responders in Non-Small cell lung cancer (NSCLC) patients with high accuracy using baseline F-18 FDG PET/CT images.
Source: Journal of Nuclear Medicine - Category: Nuclear Medicine Authors: Tags: Technical Advances Posters Source Type: research
Objective Our aim was to assess the significance of metabolic positron emission tomography (PET) parameters for the prediction of occult mediastinal lymph node metastasis (OLM) and recurrence in patients with clinical-N0 nonsmall cell lung cancer (NSCLC) after surgical resection and lymph node dissection. Materials and methods We evaluated 98 patients with NSCLC [52 adenocarcinoma (ADC), 46 squamous cell carcinoma (SQCC)] who had undergone initial/preoperative fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT). Eligibility criteria for participation were clinically staged ...
Source: Nuclear Medicine Communications - Category: Nuclear Medicine Tags: ORIGINAL ARTICLES Source Type: research
Authors: Ma W, Wang M, Li X, Huang H, Zhu Y, Song X, Dai D, Xu W Abstract The aim of the present study was to investigate the prognostic value of quantitative [18F]fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT) parameters for patients with non-small cell lung cancer (NSCLC). The present study conducted a retrospective review of the medical records of 203 patients with NSCLC, of which 193 patients underwent baseline 18F-FDG PET/CT prior to initial therapy. Multivariate analyses using Cox's proportional hazards regression were performed for the assessment of the association betwe...
Source: Oncology Letters - Category: Cancer & Oncology Tags: Oncol Lett Source Type: research
Conclusions: This proof-of-concept study preliminarily indicates the safety and efficacy of 68Ga-NOTA-3P-TATE-RGD, a novel PET tracer targeting both SSTR and integrin αvβ3, with the potential to detect NSCLC and SCLC in a single PET/CT scan using the same tracer.
Source: Journal of Nuclear Medicine - Category: Nuclear Medicine Authors: Tags: Preclinical Probes for Oncology Posters Source Type: research
207Objectives: To prospectively evaluate the role of metabolic tumor volume based semiquantitative parameters (primary metabolic tumor volume (MTV), primary tumor total lesion glycolysis(TLG), whole body MTV(MTVwb) and whole body TLG (TLGwb)) and comparison with routinely used SUV based parameters (Maximum Standardized Uptake value (SUVmax), average SUV (SUVavg), tumor to background ratio (TBR)) evaluated from baseline F18-FDG PET/CT in prediction of overall survival(OS) in biopsy proven non-small cell lung cancer(NSCLC). Material and Methods: One hundred thirty patients (108 male, 22 female) with biopsy proven NSCLC (69 -...
Source: Journal of Nuclear Medicine - Category: Nuclear Medicine Authors: Tags: Lung Cancer I: Relationship between PET-Derived Metabolic Variables, Molecular Biomarkers and Prognosis Source Type: research
Conclusions:Texture parameters evaluated from baseline F18-FDG PET/CT performed better than routinely used metabolic parameters and can be used as reliable prognostic markers of OS in NSCLC.
Source: Journal of Nuclear Medicine - Category: Nuclear Medicine Authors: Tags: Lung Cancer I: Relationship between PET-Derived Metabolic Variables, Molecular Biomarkers and Prognosis Source Type: research
281Objectives: To prospectively evaluate the role of texture parameters obtained from baseline and post-therapy F18-FDG PET/CT as well as their percentage changes in treatment response assessment of biopsy proven non-small cell lung cancer (NSCLC). Materials and Method: Thirty patients (25 male, 5 female) with biopsy proven NSCLC (16 adenocarcinoma, 8 Squamous cell carcinoma and 6 NOS) and mean age 55.83±11.72 years were enrolled in the study. Each patient underwent baseline and post-therapy whole body F-18 FDG PET/CT scan on Biograph mCT 64 slice, after 60 minutes of injecting 5.18-7.77 MBq/kg of F18-FDG intravenou...
Source: Journal of Nuclear Medicine - Category: Nuclear Medicine Authors: Tags: Lung Cancer II: PET-Derived Parameters as Predictors of Response to Therapy Source Type: research
More News: Adenocarcinoma | Cancer | Cancer & Oncology | Carcinoma | CT Scan | Lung Cancer | Non-Small Cell Lung Cancer | PET Scan | Radiology | Skin Cancer | Squamous Cell Carcinoma | Study | Training | UK Health | Universities & Medical Training