Trial Outcomes & Findings for Fujifilm Feature Comparison Reader Study to Compare Old vs New Processing (NCT NCT03586583)

NCT ID: NCT03586583

Last Updated: 2023-09-21

Results Overview

For each reader, the cases judged as non-inferior with modified processing will be calculated for each feature and each view. Non-inferiority will be defined as a score of equivalence between modified and original or a score indicating the modified image processing is superior. Readers compared several general mammographic features using the following 5-point scale: negative 2 (left image much superior) negative 1 (left image somewhat superior) 0 (left and right images are equivalent) positive 1 (right image somewhat superior) positive 2 (right image much superior)

• Recruitment status: COMPLETED
• Target enrollment: 600 participants
• Primary outcome timeframe: 1 day reading session
• Results posted on: 2023-09-21

Participant Flow

6 qualified radiologists were shown DBT (Tomosynthesis) of a breast with the modified processing side-by-side the same breast with the original processing. Each reader was given a unique sample of 100 cases to review. Images were acquired under Fujifilm image acquisition protocol FMSU2013-004A

Participant milestones

Measure	FBP (Old Processing) and ISR (New Processing) Radiologists reviewed images of a breast with ISR side-by-side the same breast with FBP. They were blinded to the image reconstruction and presentation processing methods. Radiologists compared several general mammographic features and were asked to grade each feature on each view using the 5-point Likert scale outlined previously.
Overall Study STARTED	600
Overall Study COMPLETED	600
Overall Study NOT COMPLETED	0

Reasons for withdrawal

Withdrawal data not reported

Baseline Characteristics

Fujifilm Feature Comparison Reader Study to Compare Old vs New Processing

Baseline characteristics by cohort

Measure	FBP (Old Processing)/ISR (New Processing) n=600 Participants Filtered back projection (old processing) and ISR (new processing) images of the same breast are presented side-by-side for radiologist review.
Age, Categorical <=18 years	0 Participants n=5 Participants
Age, Categorical Between 18 and 65 years	563 Participants n=5 Participants
Age, Categorical >=65 years	37 Participants n=5 Participants
Sex: Female, Male Female	600 Participants n=5 Participants
Sex: Female, Male Male	0 Participants n=5 Participants
Ethnicity (NIH/OMB) Hispanic or Latino	7 Participants n=5 Participants
Ethnicity (NIH/OMB) Not Hispanic or Latino	590 Participants n=5 Participants
Ethnicity (NIH/OMB) Unknown or Not Reported	3 Participants n=5 Participants
Region of Enrollment United States	600 participants n=5 Participants

PRIMARY outcome

Timeframe: 1 day reading session

Population: Six readers read a cohort of images acquired under a prior acquisition study protocol (FMSU2013-004A). Statistician provided results for ISR (modified/new processing only) that was already analyzed/compared to FBP (old processing).

For each reader, the cases judged as non-inferior with modified processing will be calculated for each feature and each view. Non-inferiority will be defined as a score of equivalence between modified and original or a score indicating the modified image processing is superior. Readers compared several general mammographic features using the following 5-point scale:

negative 2 (left image much superior) negative 1 (left image somewhat superior) 0 (left and right images are equivalent) positive 1 (right image somewhat superior) positive 2 (right image much superior)

Outcome measures

Measure	ISR (New Processing) n=600 Participants Iterative super resolution; new processing. ISR: New processing
Comparison of ISR (Modified Processing) to FBP (Original Processing) MLO	0.83 5-point scale Interval 0.66 to 0.95
Comparison of ISR (Modified Processing) to FBP (Original Processing) Uniformity	0.83 5-point scale Interval 0.68 to 0.95
Comparison of ISR (Modified Processing) to FBP (Original Processing) Contrast	0.80 5-point scale Interval 0.66 to 0.9
Comparison of ISR (Modified Processing) to FBP (Original Processing) Sharp	0.79 5-point scale Interval 0.67 to 0.89
Comparison of ISR (Modified Processing) to FBP (Original Processing) Skin	0.84 5-point scale Interval 0.68 to 0.96
Comparison of ISR (Modified Processing) to FBP (Original Processing) Noise	0.83 5-point scale Interval 0.69 to 0.95
Comparison of ISR (Modified Processing) to FBP (Original Processing) Artifact	0.84 5-point scale Interval 0.69 to 0.95
Comparison of ISR (Modified Processing) to FBP (Original Processing) Mass	0.63 5-point scale Interval 0.49 to 0.75
Comparison of ISR (Modified Processing) to FBP (Original Processing) Calcification	0.87 5-point scale Interval 0.78 to 0.94
Comparison of ISR (Modified Processing) to FBP (Original Processing) Arch Distortion/Asymmetry	0.64 5-point scale Interval 0.45 to 0.79

Adverse Events

FBP (Old Processing)

Serious events: 0 serious events

Other events: 0 other events

Deaths: 0 deaths

ISR (New Processing)

Serious events: 0 serious events

Other events: 0 other events

Deaths: 0 deaths

Serious adverse events

Adverse event data not reported

Other adverse events

Adverse event data not reported

Additional Information

Randy Vader

Fujifilm Medical Systems USA, Inc.

Phone: 360-356-6821

Email: randy.vader@fujifilm.com

Results disclosure agreements

• Principal investigator is a sponsor employee
• Publication restrictions are in place