Page 101 - Read Online
P. 101
Page 428 Roy et al. Art Int Surg 2024;4:427-34 https://dx.doi.org/10.20517/ais.2024.69
SURGERY: CONSIDERATIONS FOR ADOPTION AND IMPLEMENTATION
The volume and complexity of clinical data are growing rapidly across all fields of medicine. In parallel,
computational power is expanding and becoming more accessible, while human resources continue to
remain relatively stagnant and limited in healthcare . It has been predicted that artificial intelligence (AI)
[1]
[1]
and machine learning (ML) will be ubiquitous in future clinical care . Significant areas of interest that
could revolutionize care include processing of electronic health record data, image classification, and
identification of medical errors . The promising appeal to incorporate AI into clinical practice must be
[2,3]
contextualized, and the intrinsic limitations to the use of ML algorithms acknowledged. Clinical translation
of ML tools and prospective validation articles are scarce to date and a new set of considerations for
adoption and implementation have been unveiled. This article reviews the current ML landscape in
craniofacial surgery and highlights promises, challenges, and considerations for successful clinical
translation.
Current landscape of ML in craniofacial surgery
The role of ML and AI in craniofacial surgery has previously been thoroughly reviewed . In a scoping
[4-9]
review by Mak et al. (2020), the authors identified numerous craniofacial-based studies developing ML
models . ML is particularly relevant to craniofacial surgery as it is a specialty that: (1) relies on imaging for
[4]
diagnostic purposes; (2) uses standardized and universal anatomical landmarks (soft tissue and bony); (3)
benefits from three-dimensional planning and surgical navigation; and (4) has variable outcomes and
benefits from risk prediction . Thus far, published craniofacial surgery studies using AI have been
[9]
experimental and theoretical in nature, mostly relying on retrospective datasets with very limited sample
sizes and generally single-centered.
Cleft surgery
The potential benefit of integrating AI into cleft care spans numerous facets of clinical care due to the varied
presentations (from cleft lip to velopharyngeal insufficiency and dentoalveolar discrepancies) and evolution
over patients’ growth and development. A scoping review identified previously explored areas for
implementation of AI in cleft care : prediction of risk of developing cleft lip or palate, diagnosis (prenatal
[10]
cleft presence), severity of morphological deformities of nose , speech evaluation (presence of
[11]
[12]
hypernasality, assessment of intelligibility) , surgical planning (estimation of volumetric defect of alveolar
cleft), prediction of need for orthognathic surgery, and more.
Orthognathic surgery
Orthognathic surgery lands itself well to be augmented by AI, although few studies have been published
thus far. A review of possible applications demonstrated areas of interest for future studies, including :
[13]
complex diagnoses (superimposing numerous diagnostic tools for measurement of upper airways and
management of obstructive sleep apnea), common diagnoses (lateral cephalogram review), treatment
planning (taking into account the symbiotic relationship with orthodontic changes), creation of custom
dental appliances, and much more. An early study (2019) within the field of orthognathic surgery proposed
a proof of concept that AI (specifically convolutional neural networks) can be used in order to score facial
[14]
attractiveness and apparent age in orthognathic surgery patients . Assistance in diagnosis and surgical
planning has also been proposed using CT scan images and comparing discrepancies in cephalometric
measures between the AI-generated plan and the postoperative images .
[15]
Craniosynostosis and head shape difference
Head shape differences are ubiquitous, and the rarity of craniosynostosis and the clinical ramifications of
later diagnosis generate huge diagnostic importance, which can be aided by AI. A review from 2023
explored the existing literature on the topic, highlighting that most studies thus far have used two-