Diagnostic Efficacy of CNN in Predicting Intraoperative Complications and Postoperative Outcomes in SMILE

Diagnostic Efficacy of Convolutional Neural Network Based Algorithm in Predicting Intraoperative Complications and Postoperative Outcomes in Small Incision Lenticule Extraction

To evaluate the diagnostic efficiency of the neural network in predicting complications of Small Incision Lenticule Extraction in a multi-center cross-sectional study.

Study Overview

• Status: Recruiting
• Conditions: Intraoperative Complications; Postoperative Outcomes; Small-incision Lenticule Extraction (SMILE) Surgery; Deep Convolutional Neural Network
• Intervention / Treatment: Diagnostic test: AI diagnostic algorithm

Detailed Description

The primary cause of global visual impairment currently is refractive error, and Small Incision Lenticule Extraction (SMILE) using femtosecond laser for corneal stromal lenticule extraction can alter the refractive power. However, complications such as opaque bubble layer (OBL), negative pressure detachment, and black spots may arise during the SMILE laser scanning process due to individual differences in corneal characteristics, significantly affecting the normal course of surgery and postoperative recovery. Experienced docters can often predict intraoperative complications based on scan images, patient cooperation, and other factors, but the learning curve is relatively long. At present, artificial intelligence has achieved the accuracy comparable to human physicians in the interpretation of medical imaging of many different diseases.Previously, we have trained a deep convolutional neural network for predicting intraoperative complications in SMILE procedures. The current multi-center study is designed to evaluate the efficacy of the convolutional neural network based algorithm in predicting intraoperative complications and to assess its utility in the real world.

• Study Type: Observational
• Enrollment (Estimated): 1250

Contacts and Locations

• Study Contact: Name: Jian Xiong, docter; Phone Number: +86 18170906556; Email: 894040417@qq.com
• Study Contact Backup: Name: Fu Gui, docter; Phone Number: +86 13879101919; Email: 564436578@qq.com

Study Locations

• China
  • Jiangxi
    • Nanchang, Jiangxi, China, 330000
      • Recruiting
      • The Second Affiliated Hospital of Nanchang University
      • Contact: Jian Xiong, doctor; Phone Number: +86 18170906556; Email: 894040417@qq.com
      • Contact: Fu Gui, doctor; Phone Number: +86 13879101919; Email: 564436578@qq.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

Patients from clinics in different eye centers across China. Each subject must with complete surgical video recording and medical records.

Description

Inclusion Criteria:

• A condition in which the spherical equivalent refractive error of an eye is ≤-0.50 D when ocular accommodation is relaxed;
• Age ≥18 years;
• Spherical equivalent (SE) ≥-10.0D;
• Corrected distance visual acuity (CDVA) ≥16/20;
• Stable myopia for at least 2 years;
• No contact lenses wearing for at least 2 weeks.

Exclusion Criteria:

• The presence or history of eye conditions other than myopia and astigmatism, such as keratoconus or external eye injury;
• A history of eye surgery;
• The presence or history of systemic diseases.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Eyes with SMILE surgeries; Eyes with SMILE surgeries which were performed by surgeons with experiences.	Diagnostic test: AI diagnostic algorithm; The SMILE procedures collected would be assessed by the algorithm. The performance of the algorithm would be assessed, including accuracy, AUC, sensitivity and specificity.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
AUROC of convolutional neural network in predicting OBL area	The area under the receiver operating characteristic of convolutional neural network in predicting opaque bubble layer area during the SMILE surgeries	Day 0
AUROC of convolutional neural network in predicting progressive suction loss	The area under the receiver operating characteristic of convolutional neural network in predicting progressive suction loss during the SMILE surgeries	Day 0
AUROC of convolutional neural network in predicting effective optical zone	The area under the receiver operating characteristic of convolutional neural network in predicting effective optical zone after the SMILE surgeries	Day 7
AUROC of convolutional neural network in predicting postoperative refractive error	The area under the receiver operating characteristic of convolutional neural network in predicting refractive error after the SMILE surgeries	Day 7
AUROC of convolutional neural network in predicting postoperative central corneal thickness	The area under the receiver operating characteristic of convolutional neural network in predicting central corneal thickness after the SMILE surgeries	Day 7

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Sensitivity and specificity of convolutional neural network in predicting OBL area	Sensitivity and specificity of convolutional neural network in predicting opaque bubble layer area during the SMILE surgeries	Day 0
Sensitivity and specificity of convolutional neural network in predicting progressive suction loss	Sensitivity and specificity of convolutional neural network in predicting progressive suction loss during the SMILE surgeries	Day 0
Sensitivity and specificity of convolutional neural network in predicting effective optical zone	ensitivity and specificity of convolutional neural network in predicting effective optical zone after the SMILE surgeries	Day 7, Day 30, Day 90

Collaborators and Investigators

• Sponsor: Second Affiliated Hospital of Nanchang University
• Collaborators: Hangzhou Huaxia Eye Hospital; Nanchang Bright Eye Hospital

Study record dates

Study Major Dates

• Study Start (Actual): June 15, 2021
• Primary Completion (Estimated): December 1, 2027
• Study Completion (Estimated): December 1, 2027

Study Registration Dates

• First Submitted: January 3, 2024
• First Submitted That Met QC Criteria: January 3, 2024
• First Posted (Actual): January 12, 2024

Study Record Updates

• Last Update Posted (Actual): April 24, 2026
• Last Update Submitted That Met QC Criteria: April 21, 2026
• Last Verified: January 1, 2026

More Information

Terms related to this study

• Keywords: Intraoperative Complications; Postoperative Outcomes; femtosecond laser small-incision lenticule extraction; Deep Convolutional Neural Network
• Additional Relevant MeSH Terms: Pathologic Processes; Pathological Conditions, Signs and Symptoms; Behavior; Communication; Facial Expression; Nonverbal Communication; Intraoperative Complications; Smiling
• Other Study ID Numbers: [2023] No.(96)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No