AI-based diagnosis in mandibulofacial dysostosis with microcephaly using external ear shapes

Abstract

Introduction: Mandibulo-Facial Dysostosis with Microcephaly (MFDM) is a rare disease with a broad spectrum of symptoms, characterized by zygomatic and mandibular hypoplasia, microcephaly, and ear abnormalities. Here, we aimed at describing the external ear phenotype of MFDM patients, and train an Artificial Intelligence (AI)-based model to differentiate MFDM ears from non-syndromic control ears (binary classification), and from ears of the main differential diagnoses of this condition (multi-class classification): Treacher Collins (TC), Nager (NAFD) and CHARGE syndromes.

Methods: The training set contained 1,592 ear photographs, corresponding to 550 patients. We extracted 48 patients completely independent of the training set, with only one photograph per ear per patient. After a CNN-(Convolutional Neural Network) based ear detection, the images were automatically landmarked. Generalized Procrustes Analysis was then performed, along with a dimension reduction using PCA (Principal Component Analysis). The principal components were used as inputs in an eXtreme Gradient Boosting (XGBoost) model, optimized using a 5-fold cross-validation. Finally, the model was tested on an independent validation set.

Results: We trained the model on 1,592 ear photographs, corresponding to 1,296 control ears, 105 MFDM, 33 NAFD, 70 TC and 88 CHARGE syndrome ears. The model detected MFDM with an accuracy of 0.969 [0.838-0.999] (p < 0.001) and an AUC (Area Under the Curve) of 0.975 within controls (binary classification). Balanced accuracies were 0.811 [0.648-0.920] (p = 0.002) in a first multiclass design (MFDM vs. controls and differential diagnoses) and 0.813 [0.544-0.960] (p = 0.003) in a second multiclass design (MFDM vs. differential diagnoses).

Conclusion: This is the first AI-based syndrome detection model in dysmorphology based on the external ear, opening promising clinical applications both for local care and referral, and for expert centers.

Keywords: AI; MFDM; craniofacial malformation; dysmorphology; machine learning.

Figure 1

Examples of external ear photographs for each patient group: controls, mandibulo-facial dysostosis with microcephaly (MFDM), Nager type acro-facial dysostosis (NAFD), Treacher Collins (TC), and CHARGE syndromes.

Figure 2

Vectors represent distances between MFDM mean landmarks and the control mean landmarks (A). Comparison of average MFDM (red) and control (blue) ear models after Procrustes transformation (B).

Figure 3

Comparison of average MFDM (red) and the main differential diagnoses: NAFD (green) (A, B), TC (purple) (C, D) and CHARGE (yellow) (E, F), after Procrustes transformation. Vectors (A, C, E) represent distances between MFDM mean landmarks and other groups mean landmarks.

Figure 4

UMAP representations for designs №1 (A), № 2.1 (B) and № 2.2 (C), including severity and asymmetry parameters. Each color corresponds to a patient group. MFDM, Mandibulo-Facial Dysostosis with Microcephaly; NAFD, Nager type Acro-Facial Dysostosis; TC, Treacher Collins; CHARGE, Coloboma, Heart defect, Atresia choanae, Retarded growth and development, Genital hypoplasia, Ear anomalies/deafness.

Figure 5

Empirical ROC curves for designs № 1 (A), № 2.1 (B) and № 2.2 (C). MFDM, Mandibulo-Facial Dysostosis with Microcephaly; NAFD, Nager type Acro-Facial Dysostosis; TC, Treacher Collins; CHARGE, Coloboma, Heart defect, Atresia choanae, Retarded growth and development, Genital hypoplasia, Ear anomalies/deafness.

Figure 6

Case study of automatic ear-based CHARGE syndrome diagnosis (A). (B) UMAP clustering of design № 2.1; black dot: patient. (C) probability histogram with a XGboost classifier.