Evaluation of Computer-Aided Lung Nodule Detection Software in Thoracic CT for Riverain Technologies LLC

NCT02440139 | Completed Results DMC

• 1 other identifier: 458420
• Study Type: interventional
• Target: 12
• Locations: 1 country
• Sites: 1

Brief Summary

STUDY DESIGN: This is a retrospective, multi-reader, multi-case, (MRMC) randomized reader study. OBJECTIVE: Primary: The primary objective of this clinical study is to prove that a user aided with ClearRead CT InSight (CRCTI) is superior to the unaided reader for detecting actionable lung nodules. Secondary: The secondary objective of this clinical study is to prove that the reader's reading time is not significantly increased when aided with CRCTI. NUMBER OF SUBJECTS: Retrospective CT studies from approximately 300 patients will be included in the study. Approximately 100 true positive cases and 200 normal cases. NUMBER OF READERS: A reader study with at least ten (10) participating radiologists (US Board Certified) will be conducted. PRIMARY ENDPOINTS: Scores given by the radiologists with and without ClearRead CT Insight will be recorded and compared to the true status of the study-cases. The frequency of the scores for each method (Unaided, Aided) will be tabulated and LROC curves constructed along with sensitivity, specificity, PPV, NPV and clinical actions. Additionally, machine nodule detection rate and false positives per patient on normal cases will be measured. PATIENT POPULATION : The study will target approximately one hundred (100) patients whose CT nodules were shown to be cancer and two hundred (200) normal patients. The patient population will be consistent with the national lung cancer screening protocols.

Conditions

Lung Cancer

Outcome Measures

Primary Outcomes (2)

• Comparing the Detection Accuracy, Sensitivity, and Specificity of Clinically Actionable Lung Nodules

  The detection rates in accuracy, sensitivity, and specificity on clinically actionable lung nodules were computed and compared for Aim 1 (baseline) and Arm 2 (study participants were aided with ClearRead CT software) studies.

  4 months

• Comparing LROC Curve in the Detection of Clinically Actionable Lung Nodules Among Normal

  The areas under location-specific Receiver Operating Characteristic (LROC) curves were computed and compared for Aim 1 (baseline) and Aim 2 (study participants were aided with ClearRead CT software).

  4 months

Secondary Outcomes (1)

• Average Reading Time Per Case

  4 months

Study Arms (2)

Arm 1

NO INTERVENTION

Participating radiologists will perform clinical reading on \~300 cases. They will mark all locations of concern (clinically actionable nodules) on conventional thoracic CT images. Computer will measure time, readers score regions according to action and suspiciousness.

Arm 2

ACTIVE COMPARATOR

Participating radiologists will perform clinical reading on \~300 cases. They will mark all locations of concern (clinically actionable nodules) on thoracic CT images aided by ClearRead CT Insight software as the intervention. Computer will measure time, readers score regions according to action and suspiciousness.

Device: ClearRead CT Insight

Interventions

ClearRead CT Insight DEVICE

During the second reading session (concurrent read), the radiologist will be presented with a standard appearing CT with CADe marks placed and the vessel suppressed same slice with the vessel suppressed view as the intervention. The second image, vessel suppressed, will be grayed out until the radiologist move the mouse to the second panel. The radiologist will mark locations. These may or may not correspond to the locations of the CAD markers. As before, the radiologist will assign a level of suspicious to each mark and indicate the need, if any, of an additional diagnostic action (CT Follow-up, Contrast CT, PET-CT, or Biopsy).

Arm 2

Eligibility Criteria

• Age: 55 Years - 77 Years
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• A thoracic CT case that is associated or with a feature of the following:
• Asymptomatic patients age 55 - 77 with history of smoking
• Primary Lung Cancer
• Biopsy Proven (LuRADS 5) with radiology report
• Screen detected event plus 1 prior CT. 2 year of prior CTs is preferred, if available.
• \<= 3mm slice spacing, no gaps
• Use standard reconstruction kernels
• Maximum of 5 nodules per image
• Nodules must be 5-20mm in size
• With or without contrast

You may not qualify if:

• No Acute pneumothorax
• Both lungs must be fully visible within the field of view
• Apices cannot be cropped
• No excessive motion artifacts
• Symptomatic patients with co-morbidities

Sponsors & Collaborators

• Virginia Polytechnic Institute and State University: lead
• Riverain Technologies: collaborator

Study Sites (1)

Arlington Innovation Center: Health Resaerch

Arlington, Virginia, 22203, United States

Location

Related Publications (6)

• Beyer F, Zierott L, Fallenberg EM, Juergens KU, Stoeckel J, Heindel W, Wormanns D. Comparison of sensitivity and reading time for the use of computer-aided detection (CAD) of pulmonary nodules at MDCT as concurrent or second reader. Eur Radiol. 2007 Nov;17(11):2941-7. doi: 10.1007/s00330-007-0667-1. Epub 2007 May 22.

  PMID: 17929026 BACKGROUND

• Dorfman DD, Berbaum KS, Metz CE. Receiver operating characteristic rating analysis. Generalization to the population of readers and patients with the jackknife method. Invest Radiol. 1992 Sep;27(9):723-31. No abstract available.

  PMID: 1399456 BACKGROUND

• Foti G, Faccioli N, D'Onofrio M, Contro A, Milazzo T, Pozzi Mucelli R. Evaluation of a method of computer-aided detection (CAD) of pulmonary nodules with computed tomography. Radiol Med. 2010 Sep;115(6):950-61. doi: 10.1007/s11547-010-0556-6. Epub 2010 Jun 23. English, Italian.

  PMID: 20574707 BACKGROUND

• Matsumoto S, Ohno Y, Aoki T, Yamagata H, Nogami M, Matsumoto K, Yamashita Y, Sugimura K. Computer-aided detection of lung nodules on multidetector CT in concurrent-reader and second-reader modes: a comparative study. Eur J Radiol. 2013 Aug;82(8):1332-7. doi: 10.1016/j.ejrad.2013.02.005. Epub 2013 Mar 6.

  PMID: 23480965 BACKGROUND

• Obuchowski NA. New methodological tools for multiple-reader ROC studies. Radiology. 2007 Apr;243(1):10-2. doi: 10.1148/radiol.2432060387. No abstract available.

  PMID: 17392244 BACKGROUND

• Obuchowski NA, McClish DK. Sample size determination for diagnostic accuracy studies involving binormal ROC curve indices. Stat Med. 1997 Jul 15;16(13):1529-42. doi: 10.1002/(sici)1097-0258(19970715)16:133.0.co;2-h.

  PMID: 9249923 BACKGROUND

MeSH Terms

Conditions

Lung Neoplasms

Condition Hierarchy (Ancestors)

Respiratory Tract Neoplasms; Thoracic Neoplasms; Neoplasms by Site; Neoplasms; Lung Diseases; Respiratory Tract Diseases

Limitations and Caveats

This study only used a relatively low number of thoracic CT cases read by 12 radiologists. The study cases were mainly collected from the National Lung Screen Trial (NLST) lung cancer screening program. In addition, the study suffered from readers' lack of learning time of using the AI software.

Results Point of Contact

• Title: Dr. ShihChung Benedict Lo
• Organization: Virginia Tech

benlo@vt.edu

301-9215754

Study Officials

• Matthew T Freedman, M.D.

  Virginia Polytechnic Institute and State University

  STUDY DIRECTOR

Publication Agreements

• PI is Sponsor Employee: No
• Restrictive Agreement: No

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: DOUBLE
• Who Masked: PARTICIPANT, INVESTIGATOR
• Purpose: DIAGNOSTIC
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Adjunct Professor

Study Record Dates

• First Submitted: April 15, 2015
• First Posted: May 12, 2015
• Study Start: April 1, 2015
• Primary Completion: September 1, 2017
• Study Completion: March 1, 2018
• Last Updated: November 4, 2021
• Results First Posted: October 4, 2021

Record last verified: 2021-11

Data Sharing

• IPD Sharing: Will not share

Locations

US (1)