Back to results

Evaluation of Use of Diagnostic AI for Lung Cancer in Practice

Primary Purpose

Lung Cancer

Status

Unknown status

Phase

Not Applicable

Locations

Hong Kong

Study Type

Interventional

Intervention

AI-human interaction

About this trial

This is an interventional diagnostic trial for Lung Cancer

Eligibility Criteria

undefined - undefined (Child, Adult, Older Adult)All SexesAccepts Healthy Volunteers

Inclusion Criteria:

The participant performs radiology screenings professionally

Exclusion Criteria:

Sites / Locations

University of Hong Kong

Arms of the Study

Arm 1

Arm 2

Arm 3

Arm Type

Experimental

Arm Label

Probabilistic Classification

Classification Plus Detection

Classification With Delayed Detection

Arm Description

Radiologists see a "score" from 1-100 that represents the AI's prediction of whether the CT-scan comes from a patient with cancer or not before beginning their analysis of the scan.

Radiologists see a "score" from 1-100 that represents the AI's prediction of whether the CT-scan comes from a patient with cancer or not before beginning their analysis of the scan. They also see ROIs identified by the AI that represent lung nodules.

Radiologists see a "score" from 1-100 that represents the AI's prediction of whether the CT-scan comes from a patient with cancer or not before beginning their analysis of the scan. After identifying their own ROIs, the radiologist then can see ROIs identified by the AI that represent lung nodules before making final decisions.

Outcomes

Primary Outcome Measures

Classification accuracy

This compares radiologists' classifications with the ground truth in the tested cases.

Secondary Outcome Measures

detection concordance

Evaluation of concordance between radiologists in the tested cases in detection of lung nodules > 4 mm

Full Information

NCT ID

NCT03780582

First Posted

December 16, 2018

Last Updated

July 20, 2019

Sponsor

Ensemble Group Holdings, LLC

1. Study Identification

Unique Protocol Identification Number

NCT03780582

Brief Title

Evaluation of Use of Diagnostic AI for Lung Cancer in Practice

Official Title

Evaluation of Use of Diagnostic AI for Lung Cancer in Practice

Study Type

Interventional

2. Study Status

Record Verification Date

July 2019

Overall Recruitment Status

Unknown status

Study Start Date

December 14, 2018 (Actual)

Primary Completion Date

December 15, 2019 (Anticipated)

Study Completion Date

December 15, 2019 (Anticipated)

3. Sponsor/Collaborators

Responsible Party, by Official Title

Sponsor

Name of the Sponsor

Ensemble Group Holdings, LLC

4. Oversight

Studies a U.S. FDA-regulated Drug Product

Studies a U.S. FDA-regulated Device Product

Data Monitoring Committee

5. Study Description

Brief Summary

This study investigates ways of improving radiologists performance of the classification of CT-scans as cancerous or non-cancerous. Participants interact with an AI to classify CT-scans under three different conditions.

Detailed Description

The three conditions are as follows: "probabilistic classification", where the radiologist diagnoses scans using an AI cancer likelihood score; "classification plus detection", where the radiologist see detecting lung nodules in addition to the AI's probabilistic classification score before making her own examination of the CT-scan; and "classification with delayed detection", where the radiologist identifies regions of interest independently of the AI and then sees the AI's detected ROIs.

6. Conditions and Keywords

Primary Disease or Condition Being Studied in the Trial, or the Focus of the Study

Lung Cancer

7. Study Design

Primary Purpose

Diagnostic

Study Phase

Not Applicable

Interventional Study Model

Crossover Assignment

Masking

Participant

Allocation

Randomized

Enrollment

15 (Anticipated)

8. Arms, Groups, and Interventions

Arm Title

Probabilistic Classification

Arm Type

Experimental

Arm Description

Radiologists see a "score" from 1-100 that represents the AI's prediction of whether the CT-scan comes from a patient with cancer or not before beginning their analysis of the scan.

Arm Title

Classification Plus Detection

Arm Type

Experimental

Arm Description

Arm Title

Classification With Delayed Detection

Arm Type

Experimental

Arm Description

Radiologists see a "score" from 1-100 that represents the AI's prediction of whether the CT-scan comes from a patient with cancer or not before beginning their analysis of the scan. After identifying their own ROIs, the radiologist then can see ROIs identified by the AI that represent lung nodules before making final decisions.

Intervention Type

Behavioral

Intervention Name(s)

AI-human interaction

Intervention Description

Exploring what kinds of AI-human interaction improve radiologists detection accuracy.

Primary Outcome Measure Information:

Title

Classification accuracy

Description

This compares radiologists' classifications with the ground truth in the tested cases.

Time Frame

up to 4 months after initiation of evaluation of the test set

Secondary Outcome Measure Information:

Title

detection concordance

Description

Evaluation of concordance between radiologists in the tested cases in detection of lung nodules > 4 mm

Time Frame

up to 4 months after initiation of evaluation of the test set

10. Eligibility

Sex

All

Accepts Healthy Volunteers

Eligibility Criteria

Inclusion Criteria: The participant performs radiology screenings professionally Exclusion Criteria: -

Facility Information:

Facility Name

University of Hong Kong

City

Hong Kong

Country

Hong Kong

12. IPD Sharing Statement

Plan to Share IPD

Undecided

Learn more about this trial

Evaluation of Use of Diagnostic AI for Lung Cancer in Practice

We'll reach out to this number within 24 hrs