Diagnostic Efficiency of Artificial Intelligence for Surgical Neuropathology

NCT04671368 | Unknown

• 1 other identifier: PAAI2020
• Study Type: interventional
• Target: 141
• Locations: 0 countries
• Sites: N/A

Brief Summary

This is a multi-center, prospective, self-controlled, diagnostic accuracy comparative study of Artificial Intelligence Diagnostic System for Surgical Neuropathology. The investigators will compare the diagnostic efficiency of Artificial Intelligence with that of practicing pathologists, and suppose that the diagnostic efficiency of artificial intelligence in prospective clinical data is no less than that of pathologists.

Conditions

Central Nervous System Neoplasms

Keywords

Artificial Intelligence; CNS Tumor; Surgical Pathology; Diagnostic Accuracy Study

Outcome Measures

Primary Outcomes (1)

• Diagnostic Accuracy of Study Arms

  The number of correctly diagnosed participants by study arms divided by the total number of participants

  1 week after the last patient's diagnosis is completed

Secondary Outcomes (2)

• Sensitivity and specificity of Study Arms

  1 week after the last patient's diagnosis is completed

• Spearman Coefficient of Study Arms related to Gold Standard

  1 week after the last patient's diagnosis is completed

Study Arms (3)

Artificial Intelligence

EXPERIMENTAL

A deep learning based artificial intelligence diagnostic system(DOI:10.1093/neuonc/noaa163)

Diagnostic Test: Artificial Intelligence

Practicing Pathologists

ACTIVE COMPARATOR

One pathologist who has at least 5 years of experience

Diagnostic Test: Practicing Pathologists

Gold Standard

OTHER

A committee composed of two expert pathologists who has at least 10 years of experience and one expert pathologist who has at least 15 years of experience

Diagnostic Test: Gold Standard

Interventions

Artificial Intelligence DIAGNOSTIC_TEST

The investigators will use the Artificial Intelligence Diagnostic System to review the H＆E stained slide of each patient and then report the classification of the tumor on a 10-type scale.

Artificial Intelligence

Practicing Pathologists DIAGNOSTIC_TEST

The ordinary pathologist will review the H＆E stained slide of each patient(without additional informations such as: Immunohistochemistry et al.) and then report the classification of the tumor on a 10-type scale only bases on the slide images

Practicing Pathologists

Gold Standard DIAGNOSTIC_TEST

Firstly, the two expert pathologist(\>=10 years of experience) will review the H＆E stained slide of each patient on their own (with additional informations such as: Immunohistochemistry et al.) and then report the classification of the tumor on a 10-type scale.If they report the same opinion, that opinion will perform as the ground truth; while if their opinion clash with each other, the expert pathologist(\>=15 years of experience) will get involved and the agreement of three experts will perform as the ground truth

Gold Standard

Eligibility Criteria

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

You may qualify if:

• Patients or their guardians understand the research process, agree to use their data, and sign the informed consent form;
• Aged \>=18 years;
• MRI shows intracranial spaceoccupying lesions;
• The clinical diagnosis is glioma, metastasis or lymphoma thus requiring surgical treatment;
• The patient is willing to accept the surgery.

You may not qualify if:

• The patient has serious underlying diseases thus is not suitable for surgery;
• After further clinical evaluation, surgical treatment was not the best choice;
• The patient participate in clinical research of other drugs or devices;
• The researchers believe that there are other factors that will make the patients unable to complete the study.

Sponsors & Collaborators

• Jinsong Wu: lead

MeSH Terms

Conditions

Central Nervous System Neoplasms

Interventions

Artificial Intelligence

Condition Hierarchy (Ancestors)

Nervous System Neoplasms; Neoplasms by Site; Neoplasms; Nervous System Diseases

Intervention Hierarchy (Ancestors)

Algorithms; Mathematical Concepts

Study Officials

• Cuiyun Wu, Ph.D

  Huashan Hospital

  STUDY DIRECTOR

Central Study Contacts

Lei Jin, DR

CONTACT

0086-13817841756; ozlei91@126.com

Yixin Ma, BA

CONTACT

0086-18001781531; 14301050150@fudan.edu.cn

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: NON RANDOMIZED
• Masking: SINGLE
• Who Masked: OUTCOMES ASSESSOR
• Masking Details: The AI group, ordinary pathologists and gold standard group will not be informed of each other's results
• Purpose: DIAGNOSTIC
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR INVESTIGATOR
• PI Title: Chief Physician of Neurosurgery Department, Vice-director of Neurosurgery Institute, Member of Ethics Committee, Clinical Professor of Surgery

Model Details: All patients will be diagnosed by both AI and ordinary pathologists, thus performing a self-controlled study

Study Record Dates

• First Submitted: December 4, 2020
• First Posted: December 17, 2020
• Study Start: February 1, 2021
• Primary Completion: February 1, 2022
• Study Completion: February 1, 2022
• Last Updated: December 17, 2020

Record last verified: 2020-12