AUTOMATIC DETECTION OF FACIAL LOCATIONS TO MEASURE FACIAL ASYMMETRY AFTER PAEDIATRIC RADIOTHERAPY

Clea Dronne; Helena Durrant; Angela Davey; Edward Henderson; Eliana Vasquez Osorio; Marianne Aznar

doi:10.1093/neuonc/noad147.059

Abstract

Abstract AIMS Children receiving radiotherapy for head and neck tumours, such as Head and Neck Rhabdomyosarcoma (HN- RMS), often experience facial asymmetry later in life. Here, we present a convolutional neural network (CNN) that will enable the automatic detection of facial anatomical landmarks on MR images taken during follow-up examinations. This model will facilitate quantitative tracking of facial asymmetry, thus revealing the centres of growth that are most affected by radiation, with the ultimate goal of defining precise dose tolerance levels. METHOD A dataset of 61 paediatric MRI images was used to train separate models that infer the locations of four facial landmarks. These locations had previously been manually labelled by two independent observers. Before training, the images were normalised and registered to optimise model performance. The model was judged to be of acceptable accuracy if the distances between the observed landmark (gold-standard location) and the location predicted by the trained models were smaller than the inter-observer distances of each facial landmark. RESULTS The model predicted landmarks which were closer to the gold-standard location than the inter-observer locations. In detail: • Optic chiasm: 0.81 ± 0.45 mm (inter-observer 1.49 mm). • Nasion: 0.70 ± 0.62 mm (inter-observer 3.24 mm). • Left Ectoconchion: 1.22 ± 0.44 mm (inter-observer 3.58 mm). • Right Ectoconchion: 0.80 ± 0.82 mm (inter-observer 4.04 mm). CONCLUSIONS This research shows promising potential for a CNN to accurately locate head and neck landmarks in paediatric MRI images. This tool can label craniofacial landmarks in large clinical datasets quicker and more accurately than can be achieved through manual techniques.

Original language	English
Journal	Neuro-Oncology
DOIs	https://doi.org/10.1093/neuonc/noad147.059
Publication status	Published - 16 Sept 2023

Access to Document

10.1093/neuonc/noad147.059

Cite this

@article{88c0500b859745738ccac60490d01a0c,

title = "AUTOMATIC DETECTION OF FACIAL LOCATIONS TO MEASURE FACIAL ASYMMETRY AFTER PAEDIATRIC RADIOTHERAPY",

abstract = "Abstract AIMS Children receiving radiotherapy for head and neck tumours, such as Head and Neck Rhabdomyosarcoma (HN- RMS), often experience facial asymmetry later in life. Here, we present a convolutional neural network (CNN) that will enable the automatic detection of facial anatomical landmarks on MR images taken during follow-up examinations. This model will facilitate quantitative tracking of facial asymmetry, thus revealing the centres of growth that are most affected by radiation, with the ultimate goal of defining precise dose tolerance levels. METHOD A dataset of 61 paediatric MRI images was used to train separate models that infer the locations of four facial landmarks. These locations had previously been manually labelled by two independent observers. Before training, the images were normalised and registered to optimise model performance. The model was judged to be of acceptable accuracy if the distances between the observed landmark (gold-standard location) and the location predicted by the trained models were smaller than the inter-observer distances of each facial landmark. RESULTS The model predicted landmarks which were closer to the gold-standard location than the inter-observer locations. In detail: • Optic chiasm: 0.81 ± 0.45 mm (inter-observer 1.49 mm). • Nasion: 0.70 ± 0.62 mm (inter-observer 3.24 mm). • Left Ectoconchion: 1.22 ± 0.44 mm (inter-observer 3.58 mm). • Right Ectoconchion: 0.80 ± 0.82 mm (inter-observer 4.04 mm). CONCLUSIONS This research shows promising potential for a CNN to accurately locate head and neck landmarks in paediatric MRI images. This tool can label craniofacial landmarks in large clinical datasets quicker and more accurately than can be achieved through manual techniques. ",

author = "Clea Dronne and Helena Durrant and Angela Davey and Edward Henderson and Osorio, \{Eliana Vasquez\} and Marianne Aznar",

year = "2023",

month = sep,

day = "16",

doi = "10.1093/neuonc/noad147.059",

language = "English",

journal = "Neuro-Oncology",

issn = "1523-5866",

publisher = "Oxford University Press",

}

TY - JOUR

T1 - AUTOMATIC DETECTION OF FACIAL LOCATIONS TO MEASURE FACIAL ASYMMETRY AFTER PAEDIATRIC RADIOTHERAPY

AU - Dronne, Clea

AU - Durrant, Helena

AU - Davey, Angela

AU - Henderson, Edward

AU - Osorio, Eliana Vasquez

AU - Aznar, Marianne

PY - 2023/9/16

Y1 - 2023/9/16

N2 - Abstract AIMS Children receiving radiotherapy for head and neck tumours, such as Head and Neck Rhabdomyosarcoma (HN- RMS), often experience facial asymmetry later in life. Here, we present a convolutional neural network (CNN) that will enable the automatic detection of facial anatomical landmarks on MR images taken during follow-up examinations. This model will facilitate quantitative tracking of facial asymmetry, thus revealing the centres of growth that are most affected by radiation, with the ultimate goal of defining precise dose tolerance levels. METHOD A dataset of 61 paediatric MRI images was used to train separate models that infer the locations of four facial landmarks. These locations had previously been manually labelled by two independent observers. Before training, the images were normalised and registered to optimise model performance. The model was judged to be of acceptable accuracy if the distances between the observed landmark (gold-standard location) and the location predicted by the trained models were smaller than the inter-observer distances of each facial landmark. RESULTS The model predicted landmarks which were closer to the gold-standard location than the inter-observer locations. In detail: • Optic chiasm: 0.81 ± 0.45 mm (inter-observer 1.49 mm). • Nasion: 0.70 ± 0.62 mm (inter-observer 3.24 mm). • Left Ectoconchion: 1.22 ± 0.44 mm (inter-observer 3.58 mm). • Right Ectoconchion: 0.80 ± 0.82 mm (inter-observer 4.04 mm). CONCLUSIONS This research shows promising potential for a CNN to accurately locate head and neck landmarks in paediatric MRI images. This tool can label craniofacial landmarks in large clinical datasets quicker and more accurately than can be achieved through manual techniques.

AB - Abstract AIMS Children receiving radiotherapy for head and neck tumours, such as Head and Neck Rhabdomyosarcoma (HN- RMS), often experience facial asymmetry later in life. Here, we present a convolutional neural network (CNN) that will enable the automatic detection of facial anatomical landmarks on MR images taken during follow-up examinations. This model will facilitate quantitative tracking of facial asymmetry, thus revealing the centres of growth that are most affected by radiation, with the ultimate goal of defining precise dose tolerance levels. METHOD A dataset of 61 paediatric MRI images was used to train separate models that infer the locations of four facial landmarks. These locations had previously been manually labelled by two independent observers. Before training, the images were normalised and registered to optimise model performance. The model was judged to be of acceptable accuracy if the distances between the observed landmark (gold-standard location) and the location predicted by the trained models were smaller than the inter-observer distances of each facial landmark. RESULTS The model predicted landmarks which were closer to the gold-standard location than the inter-observer locations. In detail: • Optic chiasm: 0.81 ± 0.45 mm (inter-observer 1.49 mm). • Nasion: 0.70 ± 0.62 mm (inter-observer 3.24 mm). • Left Ectoconchion: 1.22 ± 0.44 mm (inter-observer 3.58 mm). • Right Ectoconchion: 0.80 ± 0.82 mm (inter-observer 4.04 mm). CONCLUSIONS This research shows promising potential for a CNN to accurately locate head and neck landmarks in paediatric MRI images. This tool can label craniofacial landmarks in large clinical datasets quicker and more accurately than can be achieved through manual techniques.

UR - http://dx.doi.org/10.1093/neuonc/noad147.059

U2 - 10.1093/neuonc/noad147.059

DO - 10.1093/neuonc/noad147.059

M3 - Meeting Abstract

SN - 1523-5866

JO - Neuro-Oncology

JF - Neuro-Oncology

ER -

AUTOMATIC DETECTION OF FACIAL LOCATIONS TO MEASURE FACIAL ASYMMETRY AFTER PAEDIATRIC RADIOTHERAPY

Abstract

Access to Document

Other files and links

Fingerprint

Cite this