A novel sagittal craniosynostosis classification system based on multi-view learning algorithm

You, Lei; Deng, Yan; Zhang, Guangming; Wang, Yanfei; Bins, Griffin P.; Runyan, Christopher M.; David, Lisa R.; Zhou, Xiaobo

doi:10.1007/s00521-022-07310-5

Cited by 5 publications

(14 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…7 This level of accuracy is on par with levels for some convolutional neural networks previously used in attempts to model craniosynostosis detection. 6,21,22 In the present study, however, we demonstrated that the PAA, an automatically calculated metric of biparietal narrowing, can be of particular value in similar ML or deep neural network models by improving sagittal craniosynostosis detection with minimal overfitting costs. Given the favorable synergy between CI and PAA for the detection of sagittal craniosynostosis, we propose that metrics of biparietal narrowing should be included in the construction of any diagnostic ML model for which sagittal craniosynostosis is an output class.…”

Section: Discussionmentioning

confidence: 58%

“…Early and accurate recognition of the cranial deformity pattern is important for optimal treatment, and the diagnosis is typically obtained via physical examination, although various confirmatory imaging models are sometimes necessary. 4,6,7 Image processing techniques that ABBREVIATIONS AI = artificial intelligence; AUC = area under the ROC curve; CI = cephalic index; ML = machine learning; PAA = posterior arc angle; ROC = receiver operating characteristic.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Incorporation of a biparietal narrowing metric to improve the ability of machine learning models to detect sagittal craniosynostosis with 2D photographs

Anderson¹,

Jungbauer²,

Leclair

et al. 2023

Neurosurgical Focus

View full text Add to dashboard Cite

OBJECTIVE Sagittal craniosynostosis is the most common form of craniosynostosis and typically results in scaphocephaly, which is characterized by biparietal narrowing, compensatory frontal bossing, and an occipital prominence. The cephalic index (CI) is a simple metric for quantifying the degree of cranial narrowing and is often used to diagnose sagittal craniosynostosis. However, patients with variant forms of sagittal craniosynostosis may present with a "normal" CI, depending on the part of the suture that is closed. As machine learning (ML) algorithms are developed to assist in the diagnosis of cranial deformities, metrics that reflect the other phenotypic features of sagittal craniosynostosis are needed. In this study the authors sought to describe the posterior arc angle (PAA), a measurement of biparietal narrowing that is obtained with 2D photographs, and elucidate the role of PAA as an adjuvant to the CI in characterizing scaphocephaly and the potential relevance of PAA in new ML model development. METHODS The authors retrospectively reviewed 1013 craniofacial patients treated during the period from 2006 to 2021. Orthogonal top-down photographs were used to calculate the CI and PAA. Distribution densities, receiver operating characteristic (ROC) curves, and chi-square analyses were used to describe the relative predictive utility of each method for sagittal craniosynostosis. RESULTS In total, 1001 patients underwent paired CI and PAA measurements and a clinical head shape diagnosis (sagittal craniosynostosis, n = 122; other cranial deformity, n = 565; normocephalic, n = 314). The area under the ROC curve (AUC) for the CI was 98.5% (95% confidence interval 97.8%–99.2%, p < 0.001), with an optimum specificity of 92.6% and sensitivity of 93.4%. The PAA had an AUC of 97.4% (95% confidence interval 96.0%–98.8%, p < 0.001) with an optimum specificity of 94.9% and sensitivity of 90.2%. In 6 of 122 (4.9%) cases of sagittal craniosynostosis, the PAA was abnormal while the CI was normal. This means that adding a PAA cutoff branch to a partition model increases the detection of sagittal craniosynostosis. CONCLUSIONS Both CI and PAA are excellent discriminators for sagittal craniosynostosis. Using an accuracy-optimized partition model, the addition of the PAA to the CI increased model sensitivity compared to using the CI alone. Using a model that incorporates both CI and PAA could assist in the early identification and treatment of sagittal craniosynostosis via automated and semiautomated algorithms that utilize tree-based ML models.

show abstract

Section: Discussionmentioning

confidence: 58%

mentioning

confidence: 99%

Incorporation of a biparietal narrowing metric to improve the ability of machine learning models to detect sagittal craniosynostosis with 2D photographs

Anderson¹,

Jungbauer²,

Leclair

et al. 2023

Neurosurgical Focus

View full text Add to dashboard Cite

show abstract

“…Despite a well-established morphology, patient evaluation remains subjective, and limited objective measurements focus on the anterior cranium. Objective measures are needed as clinician visual assessment of global and regional severity varies considerably [17‒21]. This inconsistency coupled with the absence of a tool targeting posterior anatomy limits surgeon ability to objectively discuss operative results and evaluate longevity of surgical correction.…”

Section: Discussionmentioning

confidence: 99%

A New Measure of Posterior Morphology in Sagittal Craniosynostosis: The Occipital Bullet Index

Bins,

Cull,

Layton

et al. 2023

Pediatr Neurosurg

View full text Add to dashboard Cite

Background: Sagittal craniosynostosis (SC) is associated with scaphocephaly, an elongated narrow head shape. Assessment of regional severity in the scaphocephalic head is limited by the use of serial CT imaging or complex computer programing. Three-dimensional measurements of cranial surface morphology provide a radiation-free alternative for assessing cranial shape. This study describes the creation of an Occipital Bulleting Index (OBI), a novel tool using surface morphology to assess the regional severity in patients with SC. Methods: Surface imaging from CT scans or 3D photographs of 360 individuals with sagittal craniosynostosis and 221 normocephalic individuals were compared to identify differences in morphology. Cartesian grids were created on each individual’s surface mesh using equidistant axial and sagittal planes. Area under the curve (AUC) analyses were performed to identify trends in regional morphology and create measures capturing population differences. Results: The largest differences were located in the medial regions posteriorly. Using these population trends, a measure was created maximizing AUC. The Occipital Bullet index has an AUC of 0.72 with a sensitivity of 74% and a specificity of 61%. When the Frontal Bossing Index is applied in tandem, the two have a sensitivity of 94.7% and a specificity of 93.1%. Correlation between the two scores in individuals with SC was found to be negligible with an intraclass correlation coefficient of 0.018. Severity was found to be independent of age under 24 months, sex, and imaging modality. Conclusions: This index creates a tool for differentiating control head shapes from those with sagittal craniosynostosis, and has the potential to allow for objective evaluation of the regional severity, outcomes of different surgical techniques, and tracking shape changes in individuals over time, without the need for radiation.

show abstract

“…Several other approaches use deep learning to classify craniosynostosis, yet most of them are conducted on the basis of CT imaging [25,39]. The utilization of machine learning not only holds promise for more accurate and early diagnosis but also for understanding the underlying morphological changes associated with craniosynostosis if methods of "explainable AI" can be applied.…”

Section: Discussionmentioning

confidence: 99%

The Use of Artificial Intelligence for the Classification of Craniofacial Deformities

Kuehle,

Ringwald,

Bouffleur

et al. 2023

JCM

View full text Add to dashboard Cite

Positional cranial deformities are a common finding in toddlers, yet differentiation from craniosynostosis can be challenging. The aim of this study was to train convolutional neural networks (CNNs) to classify craniofacial deformities based on 2D images generated using photogrammetry as a radiation-free imaging technique. A total of 487 patients with photogrammetry scans were included in this retrospective cohort study: children with craniosynostosis (n = 227), positional deformities (n = 206), and healthy children (n = 54). Three two-dimensional images were extracted from each photogrammetry scan. The datasets were divided into training, validation, and test sets. During the training, fine-tuned ResNet-152s were utilized. The performance was quantified using tenfold cross-validation. For the detection of craniosynostosis, sensitivity was at 0.94 with a specificity of 0.85. Regarding the differentiation of the five existing classes (trigonocephaly, scaphocephaly, positional plagiocephaly left, positional plagiocephaly right, and healthy), sensitivity ranged from 0.45 (positional plagiocephaly left) to 0.95 (scaphocephaly) and specificity ranged from 0.87 (positional plagiocephaly right) to 0.97 (scaphocephaly). We present a CNN-based approach to classify craniofacial deformities on two-dimensional images with promising results. A larger dataset would be required to identify rarer forms of craniosynostosis as well. The chosen 2D approach enables future applications for digital cameras or smartphones.

show abstract

A novel sagittal craniosynostosis classification system based on multi-view learning algorithm

Cited by 5 publications

References 22 publications

Incorporation of a biparietal narrowing metric to improve the ability of machine learning models to detect sagittal craniosynostosis with 2D photographs

Incorporation of a biparietal narrowing metric to improve the ability of machine learning models to detect sagittal craniosynostosis with 2D photographs

A New Measure of Posterior Morphology in Sagittal Craniosynostosis: The Occipital Bullet Index

The Use of Artificial Intelligence for the Classification of Craniofacial Deformities

Contact Info

Product

Resources

About