A Deep Learning Model to Predict Breast Implant Texture Types Using Ultrasonography Images: Feasibility Development Study.

BACKGROUND: Breast implants, including textured variants, have been widely used in aesthetic and reconstructive mammoplasty. However, the textured type, which is one of the shell texture types of breast implants, has been identified as a possible etiologic factor for lymphoma, specifically breast implant-associated anaplastic large cell lymphoma (BIA-ALCL). Identifying the shell texture type of the implant is critical to diagnosing BIA-ALCL. However, distinguishing the shell texture type can be difficult due to the loss of human memory and medical history. An alternative approach is to use ultrasonography, but this method also has limitations in quantitative assessment. OBJECTIVE: This study aims to determine the feasibility of using a deep learning model to classify the shell texture type of breast implants and make robust predictions from ultrasonography images from heterogeneous sources. METHODS: A total of 19,502 breast implant images were retrospectively collected from heterogeneous sources, including images captured from both Canon and GE devices, images of ruptured implants, and images without implants, as well as publicly available images. The Canon images were trained using ResNet-50. The model's performance on the Canon dataset was evaluated using stratified 5-fold cross-validation. Additionally, external validation was conducted using the GE and publicly available datasets. The area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (PRAUC) were calculated based on the contribution of the pixels with Gradient-weighted Class Activation Mapping (Grad-CAM). To identify the significant pixels for classification, we masked the pixels that contributed less than 10%, up to a maximum of 100%. To assess the model's robustness to uncertainty, Shannon entropy was calculated for 4 image groups: Canon, GE, ruptured implants, and without implants. RESULTS: The deep learning model achieved an average AUROC of 0.98 and a PRAUC of 0.88 in the Canon dataset. The model achieved an AUROC of 0.985 and a PRAUC of 0.748 for images captured with GE devices. Additionally, the model predicted an AUROC of 0.909 and a PRAUC of 0.958 for the publicly available dataset. This model maintained the PRAUC values for quantitative validation when masking up to 90% of the least-contributing pixels and the remnant pixels in breast shell layers. Furthermore, the prediction uncertainty increased in the following order: Canon (0.066), GE (0072), ruptured implants (0.371), and no implants (0.777). CONCLUSIONS: We have demonstrated the feasibility of using deep learning to predict the shell texture type of breast implants. This approach quantifies the shell texture types of breast implants, supporting the first step in the diagnosis of BIA-ALCL.

High prevalence of ALPK3 premature terminating variants in Korean hypertrophic cardiomyopathy patients.

Background: The alpha-protein kinase 3 (ALPK3) gene (OMIM: 617608) is associated with autosomal recessive familial hypertrophic cardiomyopathy-27 (CMH27, OMIM: 618052). Recently, several studies have shown that monoallelic premature terminating variants (PTVs) in ALPK3 are associated with adult-onset autosomal dominant hypertrophic cardiomyopathy (HCMP). However, these studies were performed on patient cohorts mainly from European Caucasian backgrounds. Methods: To determine if this finding is replicated in the Korean HCMP cohort, we evaluated 2,366 Korean patients with non-syndromic HCMP using exome sequencing and compared the cohort dataset with three independent population databases. Results: We observed that monoallelic PTVs in ALPK3 were also significantly enriched in Korean patients with HCMP with an odds ratio score of 10-21. Conclusions: We suggest that ALPK3 PTV carriers be considered a risk group for developing HCMP and be monitored for cardiomyopathies.

Biallelic loss of function variants in WBP4, encoding a spliceosome protein, result in a variable neurodevelopmental delay syndrome.

Over two dozen spliceosome proteins are involved in human diseases, also referred to as spliceosomopathies. WBP4 (WW Domain Binding Protein 4) is part of the early spliceosomal complex, and was not described before in the context of human pathologies. Ascertained through GeneMatcher we identified eleven patients from eight families, with a severe neurodevelopmental syndrome with variable manifestations. Clinical manifestations included hypotonia, global developmental delay, severe intellectual disability, brain abnormalities, musculoskeletal and gastrointestinal abnormalities. Genetic analysis revealed overall five different homozygous loss-of-function variants in WBP4. Immunoblotting on fibroblasts from two affected individuals with different genetic variants demonstrated complete loss of protein, and RNA sequencing analysis uncovered shared abnormal splicing patterns, including enrichment for abnormalities of the nervous system and musculoskeletal system genes, suggesting that the overlapping differentially spliced genes are related to the common phenotypes of the probands. We conclude that biallelic variants in WBP4 cause a spliceosomopathy. Further functional studies are called for better understanding of the mechanism of pathogenicity.