The Evolution of the Role of Imaging in the Diagnosis of Craniosynostosis: A Narrative Review

The Use of Artificial Intelligence for the Classification of Craniofacial Deformities

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.

3D-2D Distance Maps Conversion Enhances Classification of Craniosynostosis

OBJECTIVE

Diagnosis of craniosynostosis using photogrammetric 3D surface scans is a promising radiation-free alternative to traditional computed tomography. We propose a 3D surface scan to 2D distance map conversion enabling the usage of the first convolutional neural networks (CNNs)-based classification of craniosynostosis. Benefits of using 2D images include preserving patient anonymity, enabling data augmentation during training, and a strong under-sampling of the 3D surface with good classification performance.

METHODS

The proposed distance maps sample 2D images from 3D surface scans using a coordinate transformation, ray casting, and distance extraction. We introduce a CNN-based classification pipeline and compare our classifier to alternative approaches on a dataset of 496 patients. We investigate into low-resolution sampling, data augmentation, and attribution mapping.

RESULTS

Resnet18 outperformed alternative classifiers on our dataset with an F1-score of 0.964 and an accuracy of 98.4%. Data augmentation on 2D distance maps increased performance for all classifiers. Under-sampling allowed 256-fold computation reduction during ray casting while retaining an F1-score of 0.92. Attribution maps showed high amplitudes on the frontal head.

CONCLUSION

We demonstrated a versatile mapping approach to extract a 2D distance map from the 3D head geometry increasing classification performance, enabling data augmentation during training on 2D distance maps, and the usage of CNNs. We found that low-resolution images were sufficient for a good classification performance.

SIGNIFICANCE

Photogrammetric surface scans are a suitable craniosynostosis diagnosis tool for clinical practice. Domain transfer to computed tomography seems likely and can further contribute to reducing ionizing radiation exposure for infants.

The cranial capacity of the Saudi population measured using 3D computed tomography scans

OBJECTIVES

To measure the cranial capacity of members of the Saudi adult population across ages and genders.

METHODS

This was a retrospective cross-sectional study that used 488 Computed Tomography (CT) scans of heads (of which 275 males) to measure cranial volume. The CT slices 0.625 mm thick were uploaded using the freely available software "3D-Slicer", which then reconstructed the images and built a 3D module.

RESULTS

The mean (±SD) cranial capacity of the males was 1481.6 (±110) cm3 (range: 1241-1723 cm3), whereas the cranial capacity of the females was 1375.4 (±104) cm3 (range: 1203-1678 cm3). This study showed that the males had a mean cranial capacity that was 7% greater than that of the females in this study. The average cranial capacity of the males between the ages of 31 and 40 years was statistically significantly larger to that of the males aged 61-80 (p<0.05).

CONCLUSION

This study demonstrated that the average cranial capacity of the males was larger than that of the females. These study results can help to determine the normal cranial capacity of adults in Saudi Arabia. Further work should be carried out to aid in establishing reference data for the Saudi population.

A Radiation-Free Classification Pipeline for Craniosynostosis Using Statistical Shape Modeling

Background: Craniosynostosis is a condition caused by the premature fusion of skull sutures, leading to irregular growth patterns of the head. Three-dimensional photogrammetry is a radiation-free alternative to the diagnosis using computed tomography. While statistical shape models have been proposed to quantify head shape, no shape-model-based classification approach has been presented yet. Methods: We present a classification pipeline that enables an automated diagnosis of three types of craniosynostosis. The pipeline is based on a statistical shape model built from photogrammetric surface scans. We made the model and pathology-specific submodels publicly available, making it the first publicly available craniosynostosis-related head model, as well as the first focusing on infants younger than 1.5 years. To the best of our knowledge, we performed the largest classification study for craniosynostosis to date. Results: Our classification approach yields an accuracy of 97.8 %, comparable to other state-of-the-art methods using both computed tomography scans and stereophotogrammetry. Regarding the statistical shape model, we demonstrate that our model performs similar to other statistical shape models of the human head. Conclusion: We present a state-of-the-art shape-model-based classification approach for a radiation-free diagnosis of craniosynostosis. Our publicly available shape model enables the assessment of craniosynostosis on realistic and synthetic data.

Poor education and urgent information need for emergency physicians about rare diseases in China

BACKGROUND

Rare diseases are serious and chronic diseases that affect no more than 1 person in 2000 (in European Union criteria). Patients suffering from RDs may come to the emergency department for life-threatening symptoms, such as acute aortic dissection, intracranial haemorrhage, and severe respiratory distress. Diagnostic delay of rare disease patients is common and often caused by low rare disease awareness among physicians. The main aim of this study was to investigate Chinese emergency physicians' basic knowledge, information access and educational needs regarding rare diseases. An online questionnaire was completed by Chinese emergency physicians during January and March 2021.

METHODS AND RESULTS

A total of 539 emergency physicians, including 200 females and 339 males, responded to the questionnaire-based study. More than half of the respondents were from Tertiary A hospitals and had engaged in medical clinical work for more than 10 years. Only 4.27% of respondents correctly estimated the prevalence of rare diseases. A few respondents knew the exact number of RDs in the first official list of rare diseases in 2018. A total of 98.5% of respondents rated their knowledge about rare diseases as minimal or insufficient. Most emergency physicians preferred to obtain information through search engines instead of specialized websites on rare diseases. A lack of practice guidelines or consensus was considered the most important reason for the diagnostic delay of RD. Practice guidelines or consensus and professional websites on rare diseases are urgently needed for emergency physicians.

CONCLUSION

The investigation shows poor knowledge of rare diseases among emergency physicians. Practice guidelines and professional websites on rare diseases were the primary urgent needs for emergency physicians. Specialized RD courses should also be added to medical education.

Active Cervical Range of Motion in Babies with Positional Plagiocephaly: Analytical Cross-Sectional Study

Positional plagiocephaly (PP) is a general term describing cranial distortion from pre- or postnatal forces on the infant head. Abnormal intrauterine forces, multiple births, primiparous mothers, obstetric interventions, prematurity, male sex, excessive time lying in the supine position, and mobility restrictions of the cervical spine have been considered as the main predisposing factors. The objective was to investigate the association between the severity of PP and the active cervical rotation and to analyze the influence of predisposing factors in babies with PP. An analytical cross-sectional study was performed on 74 babies with moderate PP. Clinical and demographic data, cranial vault asymmetry, and active cervical rotation range of motion (ROM) were measured. Associations were analyzed with generalized linear models. The mean age was 16.8 ± 5.0 weeks, and 56.8% were male. A restriction in the ROM of active cervical rotation, especially to the left side, was observed. Our models showed that cranial asymmetry was related with left active cervical rotation ROM (p = 0.034) and with being transported in a pushchair (p < 0.001). Conclusions: An increased severity of PP was related with being transported in a baby pushchair and with a reduced active cervical rotation ROM toward the most restricted side.