Deep Learning for Flap Perfusion Detection (DL-FPA)
Application of a Deep Learning-Based Model for Detecting Flap Perfusion Abnormalities After First-Stage Auricular Reconstruction in Congenital Microtia Patients
The aim of this study is to develop a convolutional neural network-based image recognition model to identify flap perfusion abnormalities after first-stage auricular reconstruction in patients with congenital microtia. In addition, the study seeks to evaluate the real-world performance of this model when applied to clinical flap monitoring.
The main questions addressed are:
- Can an artificial intelligence (AI) image recognition model accurately detect flap perfusion abnormalities from photographs of the reconstructed ear?
- Compared with conventional clinical monitoring, does AI-based image recognition alter the incidence of flap-related complications?
- How accurate is AI-based detection of flap abnormalities, and how satisfied are patients and healthcare providers with this monitoring approach?
The study is divided into two phases.
- In phase I, investigators will collect postoperative photographs of reconstructed ears and relevant clinical data from patients with congenital microtia undergoing first-stage auricular reconstruction. Images will be annotated based on the clinical judgement of experienced healthcare professionals, and these data will be used to train a convolutional neural network image recognition model. The model's ability to identify flap perfusion abnormalities will then be assessed through internal validation.
- In phase II, the clinical performance of traditional monitoring methods will be compared with AI-based image recognition monitoring.
Patients admitted between July 2027 and December 2027 who meet the inclusion criteria will receive conventional clinical flap monitoring and routine follow-up and will comprise the control group.
Patients admitted between January 2028 and June 2028 who meet the inclusion criteria will receive flap monitoring using the convolutional neural network-based image recognition module in addition to standard clinical management and follow-up and will comprise the study group.
Investigators will compare postoperative complication rates, the accuracy of detecting flap perfusion abnormalities, and satisfaction among patients and healthcare providers between the two groups.
Participants will:
- Undergo flap monitoring after first-stage auricular reconstruction according to the clinical follow-up schedule;
- Provide, or cooperate in the acquisition of, reconstructed ear photographs and relevant clinical data as required by the study;
- Receive routine follow-up and health education;
- In the study group, undergo flap monitoring assisted by the image recognition module in addition to conventional monitoring.
Studienübersicht
Status
Bedingungen
Studientyp
Einschreibung (Geschätzt)
Phase
- Unzutreffend
Kontakte und Standorte
Studienkontakt
- Name: Jiaying Gao
- Telefonnummer: +86 13795315458
- E-Mail: 1254091997@qq.com
Studienorte
-
-
-
Shanghai, China, 201112
- EYE & ENT Hospital, Fudan University
-
Kontakt:
- Jing Xu
- Telefonnummer: +86 34233106
- E-Mail: xujing10126@163.com
-
-
Teilnahmekriterien
Zulassungskriterien
Studienberechtigtes Alter
Akzeptiert gesunde Freiwillige
Studienberechtigte Geschlechter
Beschreibung
Inclusion Criteria:
- Age 6-40 years;
- Typical microtia morphology consistent with the diagnostic criteria in Chinese Plastic Surgery;
- Patients after first-stage total auricular reconstruction who have not yet undergone second-stage surgery;
- The participant or family member is able to use a smartphone.
Exclusion Criteria:
- Abnormal coagulation or other relevant laboratory findings;
- Absence of normal fascial coverage on the affected ear, or presence of eczema, infection, or other local abnormalities on the affected side;
- Severe cognitive impairment or psychiatric disorders;
- Significant dysfunction of major organs;
- The patient or family member does not consent to participation in the study.
Studienplan
Wie ist die Studie aufgebaut?
Designdetails
- Hauptzweck: Diagnose
- Zuteilung: Nicht randomisiert
- Interventionsmodell: Sequenzielle Zuweisung
- Maskierung: Single
Waffen und Interventionen
Teilnehmergruppe / Arm |
Intervention / Behandlung |
|---|---|
|
Kein Eingriff: Control group
Patients with congenital microtia will undergo first-stage auricular reconstruction with standard flap assessment and routine follow-up, without AI-based remote flap image monitoring. |
|
|
Experimental: Intervention group
Participants will undergo first-stage auricular reconstruction and receive AI-assisted remote flap monitoring postoperatively, with scheduled image acquisition and flap assessment conducted according to the study protocol. |
In addition to routine outpatient follow-up, an AI image recognition-based remote intelligent monitoring system led by specialist nurses will be implemented. After discharge, patients will receive daily prompts to upload photographs of the reconstructed ear. The AI system will analyse images in real time: if classified as normal, it will provide standard care recommendations; if classified as abnormal, it will automatically alert the clinical team. Specialist nurses will manually review abnormal images and, as appropriate, request repeat imaging, intensify home care and increase monitoring frequency, or report to the supervising physician and instruct the patient to return to the hospital for further management. |
Was misst die Studie?
Primäre Ergebnismessungen
Ergebnis Maßnahme |
Maßnahmenbeschreibung |
Zeitfenster |
|---|---|---|
|
Complication rate
Zeitfenster: Assessed 3 months postoperatively
|
Complications will be identified and recorded based on outpatient review notes and follow-up records. The complication rate will be calculated as: Complication rate = (number of patients with complications after first-stage auricular reconstruction / total number of patients) × 100%. The following common postoperative complications after first-stage auricular reconstruction will be included: hematoma requiring re-hemostasis, infection, flap necrosis, cartilage exposure, and deformation of the auricular framework. |
Assessed 3 months postoperatively
|
Sekundäre Ergebnismessungen
Ergebnis Maßnahme |
Maßnahmenbeschreibung |
Zeitfenster |
|---|---|---|
|
Patient satisfaction
Zeitfenster: Assessed 3 months postoperatively
|
Patient satisfaction will be assessed using a self-designed satisfaction questionnaire comprising 15 items divided into five domains: consultation process and environment (3 items), quality of medical and nursing services (3 items), attitude of medical and nursing staff (3 items), health education (3 items), and use of the case management platform (3 items). Each item will be rated on a 3-point Likert scale from "dissatisfied" to "satisfied," scored 1-3, yielding a total score range of 15-45. The satisfaction score will be calculated as the sum of all item scores, with higher scores indicating greater patient satisfaction. |
Assessed 3 months postoperatively
|
|
System complication detection rate
Zeitfenster: Assessed at 3 months postoperatively
|
The system complication detection rate will be calculated as: System complication detection rate = (number of cases classified as abnormal by the system and clinically confirmed to have auricular reconstruction complications / total number of cases classified as abnormal by the system) × 100%. |
Assessed at 3 months postoperatively
|
Mitarbeiter und Ermittler
Sponsor
Publikationen und hilfreiche Links
Allgemeine Veröffentlichungen
- Reinisch JF, van Hovell Tot Westerflier CVA, Gould DJ, Tahiri YT. Secondary Salvage of the Unsatisfactory Microtia Reconstruction. Plast Reconstr Surg. 2020 May;145(5):1252-1261. doi: 10.1097/PRS.0000000000006766.
- Chen X, Wang X, Zhang K, Fung KM, Thai TC, Moore K, Mannel RS, Liu H, Zheng B, Qiu Y. Recent advances and clinical applications of deep learning in medical image analysis. Med Image Anal. 2022 Jul;79:102444. doi: 10.1016/j.media.2022.102444. Epub 2022 Apr 4.
- Chen X, Ma J, Zhang T. Genetics and Epigenetics in the Genesis and Development of Microtia. J Craniofac Surg. 2024 Feb 12;35(3):e261-6. doi: 10.1097/SCS.0000000000010004. Online ahead of print.
- Deng K, Dai L, Yi L, Deng C, Li X, Zhu J. Epidemiologic characteristics and time trend in the prevalence of anotia and microtia in China. Birth Defects Res A Clin Mol Teratol. 2016 Feb;106(2):88-94. doi: 10.1002/bdra.23462. Epub 2015 Dec 17.
- Fan Y, Liu W, Fan X, Niu X, Chen X. Psychosocial status of patients with unilateral and bilateral microtia before auricular reconstruction surgery. Int J Pediatr Otorhinolaryngol. 2021 Dec;151:110928. doi: 10.1016/j.ijporl.2021.110928. Epub 2021 Sep 25.
- Li Q, Zhou X, Wang Y, Qian J, Zhang Q. Auricular reconstruction of congenital microtia by using the modified Nagata method: Personal 10-Year experience with 1350 cases. J Plast Reconstr Aesthet Surg. 2018 Oct;71(10):1462-1468. doi: 10.1016/j.bjps.2018.05.036. Epub 2018 Jun 19.
- Chen ZH, Lin L, Wu CF, Li CF, Xu RH, Sun Y. Artificial intelligence for assisting cancer diagnosis and treatment in the era of precision medicine. Cancer Commun (Lond). 2021 Nov;41(11):1100-1115. doi: 10.1002/cac2.12215. Epub 2021 Oct 6.
- Lu MY, Williamson DFK, Chen TY, Chen RJ, Barbieri M, Mahmood F. Data-efficient and weakly supervised computational pathology on whole-slide images. Nat Biomed Eng. 2021 Jun;5(6):555-570. doi: 10.1038/s41551-020-00682-w. Epub 2021 Mar 1.
- van der Voort SR, Incekara F, Wijnenga MMJ, Kapsas G, Gahrmann R, Schouten JW, Nandoe Tewarie R, Lycklama GJ, De Witt Hamer PC, Eijgelaar RS, French PJ, Dubbink HJ, Vincent AJPE, Niessen WJ, van den Bent MJ, Smits M, Klein S. Combined molecular subtyping, grading, and segmentation of glioma using multi-task deep learning. Neuro Oncol. 2023 Feb 14;25(2):279-289. doi: 10.1093/neuonc/noac166.
- Blaicher W, Pokorny H, Rockenschaub S, Puhalla H, Muhlbacher F. Organ transplantation in Austria. Ann Transplant. 1996;1(3):41-4. No abstract available.
- Li J, Jiang P, An Q, Wang GG, Kong HF. Medical image identification methods: A review. Comput Biol Med. 2024 Feb;169:107777. doi: 10.1016/j.compbiomed.2023.107777. Epub 2023 Dec 5.
- Jia Z, Wang K, Li Y, Liu Z, Qin J, Yang Q. High Precision Feature Fast Extraction Strategy for Aircraft Attitude Sensor Fault Based on RepVGG and SENet Attention Mechanism. Sensors (Basel). 2022 Dec 9;22(24):9662. doi: 10.3390/s22249662.
- Koshy R, Mahmood A. Optimizing Deep CNN Architectures for Face Liveness Detection. Entropy (Basel). 2019 Apr 20;21(4):423. doi: 10.3390/e21040423.
- Zhang J, Xie Y, Wu Q, Xia Y. Medical image classification using synergic deep learning. Med Image Anal. 2019 May;54:10-19. doi: 10.1016/j.media.2019.02.010. Epub 2019 Feb 18.
- Greener JG, Kandathil SM, Moffat L, Jones DT. A guide to machine learning for biologists. Nat Rev Mol Cell Biol. 2022 Jan;23(1):40-55. doi: 10.1038/s41580-021-00407-0. Epub 2021 Sep 13.
- Fu YY, Li CL, Zhang JL, Zhang TY. Autologous cartilage microtia reconstruction: Complications and risk factors. Int J Pediatr Otorhinolaryngol. 2019 Jan;116:1-6. doi: 10.1016/j.ijporl.2018.09.035. Epub 2018 Oct 2.
Studienaufzeichnungsdaten
Haupttermine studieren
Studienbeginn (Geschätzt)
Primärer Abschluss (Geschätzt)
Studienabschluss (Geschätzt)
Studienanmeldedaten
Zuerst eingereicht
Zuerst eingereicht, das die QC-Kriterien erfüllt hat
Zuerst gepostet (Tatsächlich)
Studienaufzeichnungsaktualisierungen
Letztes Update gepostet (Tatsächlich)
Letztes eingereichtes Update, das die QC-Kriterien erfüllt
Zuletzt verifiziert
Mehr Informationen
Begriffe im Zusammenhang mit dieser Studie
Schlüsselwörter
Zusätzliche relevante MeSH-Bedingungen
Andere Studien-ID-Nummern
- Flap Recognition Model
Plan für individuelle Teilnehmerdaten (IPD)
Planen Sie, individuelle Teilnehmerdaten (IPD) zu teilen?
Beschreibung des IPD-Plans
Individual participant data will not be shared with external researchers due to patient privacy protection and institutional data policy. Aggregated study results will be reported in publications.
Arzneimittel- und Geräteinformationen, Studienunterlagen
Studiert ein von der US-amerikanischen FDA reguliertes Arzneimittelprodukt
Studiert ein von der US-amerikanischen FDA reguliertes Geräteprodukt
Diese Informationen wurden ohne Änderungen direkt von der Website clinicaltrials.gov abgerufen. Wenn Sie Ihre Studiendaten ändern, entfernen oder aktualisieren möchten, wenden Sie sich bitte an register@clinicaltrials.gov. Sobald eine Änderung auf clinicaltrials.gov implementiert wird, wird diese automatisch auch auf unserer Website aktualisiert .