Synthetic datasets for open software development in rare disease research.

BACKGROUND: Globally, researchers are working on projects aiming to enhance the availability of data for rare disease research. While data sharing remains critical, developing suitable methods is challenging due to the specific sensitivity and uniqueness of rare disease data. This creates a dilemma, as there is a lack of both methods and necessary data to create appropriate approaches initially. This work contributes to bridging this gap by providing synthetic datasets that can form the foundation for such developments. METHODS: Using a hierarchical data generation approach parameterised with publicly available statistics, we generated datasets reflecting a random sample of rare disease patients from the United States (US) population. General demographics were obtained from the US Census Bureau, while information on disease prevalence, initial diagnosis, survival rates as well as race and sex ratios were obtained from the information provided by the US Centers for Disease Control and Prevention as well as the scientific literature. The software, which we have named SynthMD, was implemented in Python as open source using libraries such as Faker for generating individual data points. RESULTS: We generated three datasets focusing on three specific rare diseases with broad impact on US citizens, as well as differences in affected genders and racial groups: Sickle Cell Disease, Cystic Fibrosis, and Duchenne Muscular Dystrophy. We present the statistics used to generate the datasets and study the statistical properties of output data. The datasets, as well as the code used to generate them, are available as Open Data and Open Source Software. CONCLUSION: The results of our work can serve as a starting point for researchers and developers working on methods and platforms that aim to improve the availability of rare disease data. Potential applications include using the datasets for testing purposes during the implementation of information systems or tailored privacy-enhancing technologies.

Automatic cochlear multimodal 3D image segmentation and analysis using atlas-model-based method.

Objectives: To propose an automated fast cochlear segmentation, length, and volume estimation method from clinical 3D multimodal images which has a potential role in the choice of cochlear implant type, surgery planning, and robotic surgeries.Methods: Two datasets from different countries were used. These datasets include 219 clinical 3D images of cochlea from 3 modalities: CT, CBCT, and MR. The datasets include different ages, genders, and types of cochlear implants. We propose an atlas-model-based method for cochlear segmentation and measurement based on high-resolution µCT model and A-value. The method was evaluated using 3D landmarks located by two experts.Results: The average error was 0.61±0.22 mm and the average time required to process an image was 5.21±0.93 seconds (P<0.001). The volume of the cochlea ranged from 73.96 mm3 to 106.97 mm3, the cochlear length ranged from 36.69 to 45.91 mm at the lateral wall and from 29.12 to 39.05 mm at the organ of Corti.Discussion: We propose a method that produces nine different automated measurements of the cochlea: volume of scala tympani, volume of scala vestibuli, central lengths of the two scalae, the scala tympani lateral wall length, and the organ of Corti length in addition to three measurements related to A-value.Conclusion: This automatic cochlear image segmentation and analysis method can help clinician process multimodal cochlear images in approximately 5 seconds using a simple computer. The proposed method is publicly available for free download as an extension for 3D Slicer software.

Automatic intra-subject registration and fusion of multimodal cochlea 3D clinical images.

BACKGROUND: The postoperative imaging assessment of Cochlear Implant (CI) patients is imperative. The main obstacle is that Magnetic Resonance imaging (MR) is contraindicated or hindered by significant artefacts in most cases with CIs. This study describes an automatic cochlear image registration and fusion method that aims to help radiologists and surgeons to process pre-and postoperative 3D multimodal imaging studies in cochlear implant (CI) patients. METHODS AND FINDINGS: We propose a new registration method, Automatic Cochlea Image Registration (ACIR-v3), which uses a stochastic quasi-Newton optimiser with a mutual information metric to find 3D rigid transform parameters for registration of preoperative and postoperative CI imaging. The method was tested against a clinical cochlear imaging dataset that contains 131 multimodal 3D imaging studies of 41 CI patients with preoperative and postoperative images. The preoperative images were MR, Multidetector Computed Tomography (MDCT) or Cone Beam Computed Tomography (CBCT) while the postoperative were CBCT. The average root mean squared error of ACIR-v3 method was 0.41 mm with a standard deviation of 0.39 mm. The results were evaluated quantitatively using the mean squared error of two 3D landmarks located manually by two neuroradiology experts in each image and compared to other previously known registration methods, e.g. Fast Preconditioner Stochastic Gradient Descent, in terms of accuracy and speed. CONCLUSIONS: Our method, ACIR-v3, produces high resolution images in the postoperative stage and allows for visualisation of the accurate anatomical details of the MRI with the absence of significant metallic artefacts. The method is implemented as an open-source plugin for 3D Slicer tool.

New ultra-fast algorithm for cochlear implant misalignment detection.

PURPOSE: Postoperative imaging following cochlear implant (CI) placement is currently the only means of diagnosing proper electrode position. Manual multiplanar reconstruction (MPR) analysis of CT and CBCT is time-consuming and requires extensive training. This study aims to evaluate the rate of CI misalignment and to determine the amount of time necessary to reach a diagnosis of correct versus incorrect CI placement for readers of different experience levels, using a novel algorithm for image analysis (ACIR) compared to MPR analysis. METHOD: The retrospective single centre study included 333 patients with cochlear implant surgery between May 2002 and May 2021. Postoperative CT and CBCT images were evaluated in three subgroups and the time to diagnosis was documented. Group 1: image evaluation using conventional MPR analysis; group 2: image evaluation by an experienced neuroradiologist via a novel ultra-fast algorithm; group 3: image evaluation by a young specialist via novel ultra-fast algorithm. T-test and Pearson's chi-squared test were used for inter-group comparisons. RESULTS: 333 patients (63.3 ± 15.9 years; 188 men) with 335 CIs were evaluated. The rate of CI misalignment diagnosed from 3D imaging was 14.3% (n = 48). MPR analysis required 255.7 ± 70.4 s per temporal bone, whereas Slicer plugin reduced analysis time to 83.3 ± 7.7 s (p < 0.001) for the experienced reader and 89.6 ± 8.7 s for the young specialist (p < 0.001). CONCLUSION: 3D postoperative imaging reveals high incidences of CI misalignment. Application of a novel ultra-fast algorithm significantly reduces the time for diagnosis compared to MPR analysis for readers of varying experience levels.