Early detection of lung cancer recurrence after stereotactic ablative radiation therapy: Radiomics system design

Salma Dammak, David Palma, Sarah Mattonen, Suresh Senan, Aaron D. Ward

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Stereotactic ablative radiotherapy (SABR) is the standard treatment recommendation for Stage I non-small cell lung cancer (NSCLC) patients who are inoperable or who refuse surgery. This option is well tolerated by even unfit patients and has a low recurrence risk post-treatment. However, SABR induces changes in the lung parenchyma that can appear similar to those of recurrence, and the difference between the two at an early follow-up time point is not easily distinguishable for an expert physician. We hypothesized that a radiomics signature derived from standard-of-care computed tomography (CT) imaging can detect cancer recurrence within six months of SABR treatment. This study reports on the design phase of our work, with external validation planned in future work. In this study, we performed cross-validation experiments with four feature selection approaches and seven classifiers on an 81-patient data set. We extracted 104 radiomics features from the consolidative and the peri-consolidative regions on the follow-up CT scans. The best results were achieved using the sum of estimated Mahalanobis distances (Maha) for supervised forward feature selection and a trainable automatic radial basis support vector classifier (RBSVC). This system produced an area under the receiver operating characteristic curve (AUC) of 0.84, an error rate of 16.4%, a false negative rate of 12.7%, and a false positive rate of 20.0% for leaveone patient out cross-validation. This suggests that once validated on an external data set, radiomics could reliably detect post-SABR recurrence and form the basis of a tool assisting physicians in making salvage treatment decisions.

Original languageEnglish
Title of host publicationMedical Imaging 2018
Subtitle of host publicationComputer-Aided Diagnosis
PublisherSPIE
Volume10575
ISBN (Electronic)9781510616394
DOIs
Publication statusPublished - 1 Jan 2018
EventMedical Imaging 2018: Computer-Aided Diagnosis - Houston, United States
Duration: 12 Feb 201815 Feb 2018

Conference

ConferenceMedical Imaging 2018: Computer-Aided Diagnosis
CountryUnited States
CityHouston
Period12/02/201815/02/2018

Cite this

Dammak, S., Palma, D., Mattonen, S., Senan, S., & Ward, A. D. (2018). Early detection of lung cancer recurrence after stereotactic ablative radiation therapy: Radiomics system design. In Medical Imaging 2018: Computer-Aided Diagnosis (Vol. 10575). [1057503] SPIE. https://doi.org/10.1117/12.2292444
Dammak, Salma ; Palma, David ; Mattonen, Sarah ; Senan, Suresh ; Ward, Aaron D. / Early detection of lung cancer recurrence after stereotactic ablative radiation therapy : Radiomics system design. Medical Imaging 2018: Computer-Aided Diagnosis. Vol. 10575 SPIE, 2018.
@inproceedings{24ec7db007a5487b8b0120b19a423139,
title = "Early detection of lung cancer recurrence after stereotactic ablative radiation therapy: Radiomics system design",
abstract = "Stereotactic ablative radiotherapy (SABR) is the standard treatment recommendation for Stage I non-small cell lung cancer (NSCLC) patients who are inoperable or who refuse surgery. This option is well tolerated by even unfit patients and has a low recurrence risk post-treatment. However, SABR induces changes in the lung parenchyma that can appear similar to those of recurrence, and the difference between the two at an early follow-up time point is not easily distinguishable for an expert physician. We hypothesized that a radiomics signature derived from standard-of-care computed tomography (CT) imaging can detect cancer recurrence within six months of SABR treatment. This study reports on the design phase of our work, with external validation planned in future work. In this study, we performed cross-validation experiments with four feature selection approaches and seven classifiers on an 81-patient data set. We extracted 104 radiomics features from the consolidative and the peri-consolidative regions on the follow-up CT scans. The best results were achieved using the sum of estimated Mahalanobis distances (Maha) for supervised forward feature selection and a trainable automatic radial basis support vector classifier (RBSVC). This system produced an area under the receiver operating characteristic curve (AUC) of 0.84, an error rate of 16.4{\%}, a false negative rate of 12.7{\%}, and a false positive rate of 20.0{\%} for leaveone patient out cross-validation. This suggests that once validated on an external data set, radiomics could reliably detect post-SABR recurrence and form the basis of a tool assisting physicians in making salvage treatment decisions.",
keywords = "computed tomography imaging, computer-aided diagnosis, early detection of recurrence, lung CT feature extraction, machine learning, radiationinduced lung injury, radiomics, stage I non-small cell lung cancer, Stereotactic ablative radiation therapy",
author = "Salma Dammak and David Palma and Sarah Mattonen and Suresh Senan and Ward, {Aaron D.}",
year = "2018",
month = "1",
day = "1",
doi = "10.1117/12.2292444",
language = "English",
volume = "10575",
booktitle = "Medical Imaging 2018",
publisher = "SPIE",
address = "United States",

}

Dammak, S, Palma, D, Mattonen, S, Senan, S & Ward, AD 2018, Early detection of lung cancer recurrence after stereotactic ablative radiation therapy: Radiomics system design. in Medical Imaging 2018: Computer-Aided Diagnosis. vol. 10575, 1057503, SPIE, Medical Imaging 2018: Computer-Aided Diagnosis, Houston, United States, 12/02/2018. https://doi.org/10.1117/12.2292444

Early detection of lung cancer recurrence after stereotactic ablative radiation therapy : Radiomics system design. / Dammak, Salma; Palma, David; Mattonen, Sarah; Senan, Suresh; Ward, Aaron D.

Medical Imaging 2018: Computer-Aided Diagnosis. Vol. 10575 SPIE, 2018. 1057503.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Early detection of lung cancer recurrence after stereotactic ablative radiation therapy

T2 - Radiomics system design

AU - Dammak, Salma

AU - Palma, David

AU - Mattonen, Sarah

AU - Senan, Suresh

AU - Ward, Aaron D.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Stereotactic ablative radiotherapy (SABR) is the standard treatment recommendation for Stage I non-small cell lung cancer (NSCLC) patients who are inoperable or who refuse surgery. This option is well tolerated by even unfit patients and has a low recurrence risk post-treatment. However, SABR induces changes in the lung parenchyma that can appear similar to those of recurrence, and the difference between the two at an early follow-up time point is not easily distinguishable for an expert physician. We hypothesized that a radiomics signature derived from standard-of-care computed tomography (CT) imaging can detect cancer recurrence within six months of SABR treatment. This study reports on the design phase of our work, with external validation planned in future work. In this study, we performed cross-validation experiments with four feature selection approaches and seven classifiers on an 81-patient data set. We extracted 104 radiomics features from the consolidative and the peri-consolidative regions on the follow-up CT scans. The best results were achieved using the sum of estimated Mahalanobis distances (Maha) for supervised forward feature selection and a trainable automatic radial basis support vector classifier (RBSVC). This system produced an area under the receiver operating characteristic curve (AUC) of 0.84, an error rate of 16.4%, a false negative rate of 12.7%, and a false positive rate of 20.0% for leaveone patient out cross-validation. This suggests that once validated on an external data set, radiomics could reliably detect post-SABR recurrence and form the basis of a tool assisting physicians in making salvage treatment decisions.

AB - Stereotactic ablative radiotherapy (SABR) is the standard treatment recommendation for Stage I non-small cell lung cancer (NSCLC) patients who are inoperable or who refuse surgery. This option is well tolerated by even unfit patients and has a low recurrence risk post-treatment. However, SABR induces changes in the lung parenchyma that can appear similar to those of recurrence, and the difference between the two at an early follow-up time point is not easily distinguishable for an expert physician. We hypothesized that a radiomics signature derived from standard-of-care computed tomography (CT) imaging can detect cancer recurrence within six months of SABR treatment. This study reports on the design phase of our work, with external validation planned in future work. In this study, we performed cross-validation experiments with four feature selection approaches and seven classifiers on an 81-patient data set. We extracted 104 radiomics features from the consolidative and the peri-consolidative regions on the follow-up CT scans. The best results were achieved using the sum of estimated Mahalanobis distances (Maha) for supervised forward feature selection and a trainable automatic radial basis support vector classifier (RBSVC). This system produced an area under the receiver operating characteristic curve (AUC) of 0.84, an error rate of 16.4%, a false negative rate of 12.7%, and a false positive rate of 20.0% for leaveone patient out cross-validation. This suggests that once validated on an external data set, radiomics could reliably detect post-SABR recurrence and form the basis of a tool assisting physicians in making salvage treatment decisions.

KW - computed tomography imaging

KW - computer-aided diagnosis

KW - early detection of recurrence

KW - lung CT feature extraction

KW - machine learning

KW - radiationinduced lung injury

KW - radiomics

KW - stage I non-small cell lung cancer

KW - Stereotactic ablative radiation therapy

UR - http://www.scopus.com/inward/record.url?scp=85046251715&partnerID=8YFLogxK

U2 - 10.1117/12.2292444

DO - 10.1117/12.2292444

M3 - Conference contribution

VL - 10575

BT - Medical Imaging 2018

PB - SPIE

ER -

Dammak S, Palma D, Mattonen S, Senan S, Ward AD. Early detection of lung cancer recurrence after stereotactic ablative radiation therapy: Radiomics system design. In Medical Imaging 2018: Computer-Aided Diagnosis. Vol. 10575. SPIE. 2018. 1057503 https://doi.org/10.1117/12.2292444