Lung cancer prediction using machine learning and advanced imaging techniques

Published on Jun 1, 2018in Translational lung cancer research5.132
· DOI :10.21037/TLCR.2018.05.15
Timor Kadir22
Estimated H-index: 22
Fergus V. Gleeson58
Estimated H-index: 58
(University of Oxford)
Machine learning based lung cancer prediction models have been proposed to assist clinicians in managing incidental or screen detected indeterminate pulmonary nodules. Such systems may be able to reduce variability in nodule classification, improve decision making and ultimately reduce the number of benign nodules that are needlessly followed or worked-up. In this article, we provide an overview of the main lung cancer prediction approaches proposed to date and highlight some of their relative strengths and weaknesses. We discuss some of the challenges in the development and validation of such techniques and outline the path to clinical adoption.
📖 Papers frequently viewed together
11 Authors (Wookjin Choi, ..., Wei Lu)
54 Citations
4 Authors (Nenad Filipovic, ..., Akira Tsuda)
#1Arnaud Arindra Adiyoso Setio (Radboud University Nijmegen)H-Index: 12
#2Alberto Traverso (INFN: Istituto Nazionale di Fisica Nucleare)H-Index: 10
Last. Colin Jacobs (Radboud University Nijmegen)H-Index: 24
view all 32 authors...
Automatic detection of pulmonary nodules in thoracic computed tomography (CT) scans has been an active area of research for the last two decades. However, there have only been few studies that provide a comparative performance evaluation of different systems on a common database. We have therefore set up the LUNA16 challenge, an objective evaluation framework for automatic nodule detection algorithms using the largest publicly available reference database of chest CT scans, the LIDC-IDRI data se...
399 CitationsSource
#1Francesco Ciompi (Radboud University Nijmegen)H-Index: 23
#2Kaman Chung (Radboud University Nijmegen)H-Index: 9
Last. Bram van Ginneken (Radboud University Nijmegen)H-Index: 12
view all 13 authors...
The introduction of lung cancer screening programs will produce an unprecedented amount of chest CT scans in the near future, which radiologists will have to read in order to decide on a patient follow-up strategy. According to the current guidelines, the workup of screen-detected nodules strongly relies on nodule size and nodule type. In this paper, we present a deep learning system based on multi-stream multi-scale convolutional networks, which automatically classifies all nodule types relevan...
172 CitationsSource
The large number of indeterminate pulmonary nodules encountered incidentally or during CT-based lung screening provides considerable diagnostic and management challenges. Conventional nodule evaluation relies on visually identifiable discriminators such as size and speculation. These visible nodule features are however small in number and subject to considerable interpretation variability. With the development of novel targeted therapies for lung cancer the diagnosis and characterization of earl...
69 CitationsSource
#1Samuel G. Armato (U of C: University of Chicago)H-Index: 40
#2Karen Drukker (U of C: University of Chicago)H-Index: 30
Last. Laurence P. ClarkeH-Index: 19
view all 11 authors...
The purpose of this work is to describe the LUNGx Challenge for the computerized classification of lung nodules on diagnostic computed tomography (CT) scans as benign or malignant and report the performance of participants’ computerized methods along with that of six radiologists who participated in an observer study performing the same Challenge task on the same dataset. The Challenge provided sets of calibration and testing scans, established a performance assessment process, and created an in...
48 CitationsSource
#1Anastasia Chalkidou ('KCL': King's College London)H-Index: 9
#2Michael O'Doherty ('KCL': King's College London)H-Index: 69
Last. Paul Marsden ('KCL': King's College London)H-Index: 56
view all 3 authors...
Purpose A number of recent publications have proposed that a family of image-derived indices, called texture features, can predict clinical outcome in patients with cancer. However, the investigation of multiple indices on a single data set can lead to significant inflation of type-I errors. We report a systematic review of the type-I error inflation in such studies and review the evidence regarding associations between patient outcome and texture features derived from positron emission tomograp...
198 CitationsSource
#1Paul F. PinskyH-Index: 61
#2David S. Gierada (WashU: Washington University in St. Louis)H-Index: 43
Last. Ella A. Kazerooni (UM: University of Michigan)H-Index: 87
view all 7 authors...
The U.S. Preventive Services Task Force recently recommended (grade B) lung cancer screening with low-dose computed tomography (LDCT) for high-risk current and former smokers (1). The primary evidence used by the Task Force was the National Lung Screening Trial (NLST), which reported a 20% reduction in lung cancer–specific death associated with LDCT screening (2). Important considerations for widespread use of LDCT lung cancer screening in clinical practice include the definition of a positive r...
237 CitationsSource
#2David R BaldwinH-Index: 43
Last. I WoolhouseH-Index: 6
view all 15 authors...
135 Citations
#1Stephen A. DeppenH-Index: 18
#2Jeffrey D. Blume (Vandy: Vanderbilt University)H-Index: 31
Last. Eric L. GroganH-Index: 21
view all 14 authors...
Background Existing predictive models for lung cancer focus on improving screening or referral for biopsy in general medical populations. A predictive model calibrated for use during preoperative evaluation of suspicious lung lesions is needed to reduce unnecessary operations for a benign disease. A clinical prediction model (Thoracic Research Evaluation And Treatment [TREAT]) is proposed for this purpose. Methods We developed and internally validated a clinical prediction model for lung cancer ...
29 CitationsSource
#1Hugo J.W.L. Aerts (Brigham and Women's Hospital)H-Index: 65
#2Emmanuel Rios Velazquez (Brigham and Women's Hospital)H-Index: 13
Last. Philippe Lambin (UM: Maastricht University)H-Index: 113
view all 18 authors...
Human cancers exhibit strong phenotypic differences that can be visualized noninvasively by medical imaging. Radiomics refers to the comprehensive quantification of tumour phenotypes by applying a large number of quantitative image features. Here we present a radiomic analysis of 440 features quantifying tumour image intensity, shape and texture, which are extracted from computed tomography data of 1,019 patients with lung or head-and-neck cancer. We find that a large number of radiomic features...
2,200 CitationsSource
#1A. Mcwilliams (Vancouver General Hospital)H-Index: 2
#2Martin C. TammemagiH-Index: 29
Last. Stephen LamH-Index: 86
view all 33 authors...
A B S T R AC T Background Major issues in the implementation of screening for lung cancer by means of lowdose computed tomography (CT) are the definition of a positive result and the management of lung nodules detected on the scans. We conducted a populationbased prospective study to determine factors predicting the probability that lung nodules detected on the first screening low-dose CT scans are malignant or will be found to be malignant on follow-up. Methods We analyzed data from two cohorts...
648 CitationsSource
Cited By29
#1Bhagyashree Shah (CSU: Charles Sturt University)
#2Abeer AlsadoonH-Index: 12
Last. Azam Beg (College of Information Technology)H-Index: 10
view all 5 authors...
Deep learning (DL) is a type of machine learning capable of processing large quantities of data to provide analytic results based on a particular framework’s parameters and aims. DL is widely used in a variety of fields, including medicine. Currently, there are various DL-based prediction models for predicting cancer probability and survival. However, the specific problem is that no integrated system can predict cancer survival, probability, and presence in the medical patient’s samples. Therefo...
#1R. Sujitha (National Engineering College)H-Index: 1
#2V. Seenivasagam (National Engineering College)H-Index: 1
With the fast pace in collating big data healthcare framework and accurate prediction in detection of lung cancer at early stages, machine learning gives the best of both worlds. In this paper, a streamlining of machine learning algorithms together with apache spark designs an architecture for effective classification of images and stages of lung cancer to the greatest extent. We experiment on a combination of binary classification (SVM-non linear SVM with Radial Basis Function RBF) and Multi-cl...
5 CitationsSource
#1José Lucas Leite Calheiros (UFAL: Federal University of Alagoas)
#2Lucas Amorim (UFAL: Federal University of Alagoas)H-Index: 2
Last. Marcelo Costa Oliveira (UFAL: Federal University of Alagoas)H-Index: 8
view all 6 authors...
Lung cancer is the most lethal malignant neoplasm worldwide, with an annual estimated rate of 1.8 million deaths. Computed tomography has been widely used to diagnose and detect lung cancer, but its diagnosis remains an intricate and challenging work, even for experienced radiologists. Computer-aided diagnosis tools and radiomics tools have provided support to the radiologist’s decision, acting as a second opinion. The main focus of these tools has been to analyze the intranodular zone; neverthe...
#1Lal Hussain (University of Azad Jammu and Kashmir)H-Index: 13
#2Majid AlmaraashiH-Index: 6
Last. Saif-ur-Rehman Saif Abbasi (IIUI: International Islamic University, Islamabad)H-Index: 1
view all 5 authors...
#1Susan Fernandes (Edin.: University of Edinburgh)
#2Gareth O. S. Williams (Edin.: University of Edinburgh)H-Index: 5
Last. Kevin Dhaliwal (Edin.: University of Edinburgh)H-Index: 24
view all 10 authors...
Solitary pulmonary nodules (SPNs) are a clinical challenge, given there is no single clinical sign or radiological feature that definitively identifies a benign from a malignant SPN. The early detection of lung cancer has a huge impact on survival outcome. Consequently, there is great interest in the prompt diagnosis, and treatment of malignant SPNs. Current diagnostic pathways involve endobronchial/transthoracic tissue biopsies or radiological surveillance, which can be associated with suboptim...
#1Scott J. Adams (U of S: University of Saskatchewan)H-Index: 8
#2Prosanta Mondal (U of S: University of Saskatchewan)H-Index: 8
Last. Paul Babyn (U of S: University of Saskatchewan)H-Index: 11
view all 6 authors...
Abstract Objectives To develop a lung nodule management strategy combining the Lung CT Screening Reporting and Data System (Lung-RADS) with an artificial intelligence (AI) malignancy risk score and determine its impact on follow-up investigations and associated costs in a baseline lung cancer screening population. Materials and Methods Secondary analysis was undertaken of a data set consisting of AI malignancy risk scores and Lung-RADS classifications from six radiologists for 192 baseline low-d...
#1Maria Tsakok (John Radcliffe Hospital)H-Index: 7
#1Maria Tsakok (John Radcliffe Hospital)H-Index: 2
Last. Fergus Gleeson (John Radcliffe Hospital)H-Index: 6
view all 6 authors...
Abstract Purpose To determine how implementation of an artificial intelligence nodule algorithm, the Lung Cancer Prediction Convolutional Neural Network (LCP-CNN), at the point of incidental nodule detection would have influenced further investigation and management using a series of threshold scores at both the benign and malignant end of the spectrum. Method An observational retrospective study was performed in the assessment of nodules between 5-15 mm (158 benign, 32 malignant) detected on CT...
Lung cancer is one of the most common cancers in all over the world. It is a leading cause of cancer death in men and women in the United States. Lung cancer detection is a challenging problem in the world due to structure of cancer cell, where most of the cells overlap each other. The history of the patient and his/her histological classification in terms of lung cancer provide critical information regarding the characteristics of tissues and anatomical locations. Many of the machine learning t...
view all 3 authors...