Deep-learning Based Classification of Spine CT | NCT03790930

Studies
Deep-learning Based Classification of Spine CT

NCT03790930

Sponsor

Shanghai 10th People's Hospital

Status

Unknown status

Last Update Posted

May 12, 2020Actual

Enrollment

500Estimated

Phase

Not provided

Conditions

Surgical Procedure, Unspecified

Interventions

deep learning

Countries

China

Protocol Section

Identification Module Status Module Sponsor/Collaborators Module Oversight Module Description Module Conditions Module Design Module Arms/Interventions Module Outcomes Module Eligibility Module Contacts/Locations Module References Module IPD Sharing Statement Module

Results Section

Participant Flow Module Baseline Characteristics Module Outcome Measures Module Adverse Events Module More Info Module

Annotation Section

Annotation Module

Document Section

Large Document Module

Derived Section

Miscellaneous Info Module Condition Browse Module Intervention Browse Module

Protocol Section

Identification Module

NCT ID

Results Section

No data available

No data is available for this block.

Annotation Section

No data available

No data is available for this block.

Document Section

No data available

No data is available for this block.

Derived Section

Miscellaneous Info Module

Version Holder

Status Module

Record Verification Date

May 2020

Overall Recruitment Status or Expanded Access Status

Unknown status

Last Known Status

Recruiting

Delayed Posting

Not provided

Why Stopped

Not provided

Expanded Access Info

Start Date

Feb 22, 2019Actual

Primary Completion Date

May 2020Estimated

Completion Date

May 2020Estimated

First Submitted Date

Nov 16, 2018

First Submission Date that Met QC Criteria

Dec 29, 2018

First Posted Date

Jan 2, 2019Actual

Results Waived

Not provided

Results First Submitted Date

Not provided

Results First Submitted that Met QC Criteria

Not provided

Results First Posted Date

Not provided

Certification/Extension (aka Delayed Results) First Submitted Date

Not provided

Certification/Extension First Submitted that Passed QC Review

Not provided

Certification/Extension First Posted Date

Not provided

Last Update Submitted Date

May 10, 2020

Last Update Posted Date

May 12, 2020Actual

Sponsor/Collaborators Module

Responsible Party, by Official Title

Shisheng He, MD, Executive Director of Orthopedic Department, Shanghai 10th People's HospitalPrincipal Investigator

Lead Sponsor

Shanghai 10th People's HospitalOTHER

Collaborators

Name	Class
Third Affiliated Hospital, Sun Yat-Sen University	OTHER

Oversight Module

Has Data Monitoring Committee (DMC)

Is FDA Regulated Drug

Is FDA Regulated Device

Is Unapproved Device

Not provided

Pediatric Postmarket Surveillance of a Device Product

Not provided

Product Exported from US

Not provided

FDAAA801 Violation

Not provided

Description Module

Brief Summary

It is time-consuming for spine surgeons or radiologists to conduct manual classifications of spinal CT, which may also be correlated with high inter-observer variance. With the development of computer science, deep learning has emerged as a promising technique to classify images from individual level to pixel level. The main of the study is to automatically identify and classify the lesions, or segment targeted structures on spinal CT with deep learning.

Detailed Description

Computer tomography (CT) is one of the most important imaging tool to assist the diagnostic and treatment of spinal disease. Classification of specific targets (e.g. individuals, lesions, etc.) is one of the most common mission of medical image analysis. However, it is time-consuming for spine surgeons or radiologists to conduct manual classifications of spinal CT, which may also be correlated with high inter-observer variance. With the development of computer science, deep learning has emerged as a promising technique to classify images from individual level to pixel level. The main of the study is to automatically identify and classify the lesions, or segment targeted structures on spinal CT with deep learning.

Design Module

Study Type

Observational

Number of References to an Expanded Access Study

Not provided

Expanded Access Types

Not provided

Patient Registry

Target Follow-Up Duration

Not provided

Phases

Not provided

Interventional Study Design

Allocation

Not provided

Intervention Model

Biospecimen

No data available

No data is available for this block.

Enrollment

500Estimated

Arms/Interventions Module

Arm Groups

Label	Type	Description	Intervention Names
thin layer CT		Thin-layer CT will be manually labeled and used to train, validate and test deep learning algorithm.	Diagnostic Test: deep learning

Interventions

Name	Type	Description	Arm Group Labels	Other Names
deep learning	Diagnostic Test	manually labeled samples will be used to train, validate and test deep learning algorithm, and then realize automatic classification.	thin layer CT

Outcomes Module

Primary Outcomes

Measure	Description	Time Frame
classification accuracy	classification accuracy (e.g. area under the curve, etc.)	1 day
segmentation accuracy	segmentation accuracy of multiple structures (e.g. Dice score, etc.)	1 day

Secondary Outcomes

Not provided

Other Outcomes

Not provided

Eligibility Module

Eligibility Criteria

Inclusion Criteria:

- spinal thin layer CT

Exclusion Critera:

medals or other implants induce artifact
poor image quality

Accepts Healthy Volunteers

Sex

All

Sex/Gender Based

Not provided

Sex/Gender Description

Not provided

Minimum Age

18 Years

Maximum Age

65 Years

Standard Ages

AdultOlder Adult

Study Population

patients with thin layer spinal CT covering targeted level will be included.

Sampling Method

Non-Probability Sample

Contacts/Locations Module

Central Contacts

Name	Role	Phone	Extension	Email
Guoxin Fan	Contact	008602166307580		gfan@tongji.edu.cn

Overall Officials

Name	Affiliation	Role
Shisheng He, M.D.	Shanghai 10th People's Hospital	Principal Investigator

Locations

Facility	Status	City	State	ZIP	Country	Contacts
Shanghai Tenth People's Hospital	Recruiting	Shanghai	Shanghai Municipality	200072	China

IPD Sharing Statement Module

Plan to Share IPD

Undecided

Description

Not provided

Types

Not provided

Time Frame

Not provided

Access Criteria

Not provided

URL

Not provided

Intervention Browse Module

MeSH Terms

ID	Term
D000077321	Deep Learning

Ancestor Terms

ID	Term
D000069550	Machine Learning
D001185	Artificial Intelligence
D000465	Algorithms
D055641	Mathematical Concepts

Browse Leaves

Not provided

Browse Branches

Not provided