An automatic and accurate deep learning-based neuroimaging pipeline for the neonatal brain.

BACKGROUND: Accurate segmentation of neonatal brain tissues and structures is crucial for studying normal development and diagnosing early neurodevelopmental disorders. However, there is a lack of an end-to-end pipeline for automated segmentation and imaging analysis of the normal and abnormal neonatal brain. OBJECTIVE: To develop and validate a deep learning-based pipeline for neonatal brain segmentation and analysis of structural magnetic resonance images (MRI). MATERIALS AND METHODS: Two cohorts were enrolled in the study, including cohort 1 (582 neonates from the developing Human Connectome Project) and cohort 2 (37 neonates imaged using a 3.0-tesla MRI scanner in our hospital).We developed a deep leaning-based architecture capable of brain segmentation into 9 tissues and 87 structures. Then, extensive validations were performed for accuracy, effectiveness, robustness and generality of the pipeline. Furthermore, regional volume and cortical surface estimation were measured through in-house bash script implemented in FSL (Oxford Centre for Functional MRI of the Brain Software Library) to ensure reliability of the pipeline. Dice similarity score (DSC), the 95th percentile Hausdorff distance (H95) and intraclass correlation coefficient (ICC) were calculated to assess the quality of our pipeline. Finally, we finetuned and validated our pipeline on 2-dimensional thick-slice MRI in cohorts 1 and 2. RESULTS: The deep learning-based model showed excellent performance for neonatal brain tissue and structural segmentation, with the best DSC and the 95th percentile Hausdorff distance (H95) of 0.96 and 0.99 mm, respectively. In terms of regional volume and cortical surface analysis, our model showed good agreement with ground truth. The ICC values for the regional volume were all above 0.80. Considering the thick-slice image pipeline, the same trend was observed for brain segmentation and analysis. The best DSC and H95 were 0.92 and 3.00 mm, respectively. The regional volumes and surface curvature had ICC values just below 0.80. CONCLUSIONS: We propose an automatic, accurate, stable and reliable pipeline for neonatal brain segmentation and analysis from thin and thick structural MRI. The external validation showed very good reproducibility of the pipeline.

Comparative transcriptome analysis of the hyperaccumulator plant Phytolacca americana in response to cadmium stress.

To study the molecular mechanism of the hyperaccumulator plant Phytolacca americana against cadmium (Cd) stress, the leaves of P. americana treated with 400 µM Cd for 0, 2, 12, and 24 h were harvested for comparative transcriptome analysis. In total, 110.07 Gb of clean data were obtained, and 63,957 unigenes were acquired after being assembled. Due to the lack of P. americana genome information, only 24,517 unigenes were annotated by public databases. After Cd treatment, 5054 differentially expressed genes (DEGs) were identified. KEGG pathway enrichment analysis of DEGs showed that genes involved in the flavonoid biosynthesis and antenna proteins of photosynthesis were significantly down-regulated, while genes related to the lignin biosynthesis pathway were remarkably up-regulated, indicating that P. americana could synthesize more lignin to cope with Cd stress. Moreover, genes related to heavy metal accumulation, sulfur metabolism and glutathione metabolism were also significantly up-regulated. The gene expression pattern of several key genes related to distinct metabolic pathways was verified by qRT-PCR. The results indicated that the immobilization of lignin in cell wall, chelation, vacuolar compartmentalization, as well as the increase of thiol compounds content may be the important mechanisms of Cd detoxification in hyperaccumulator plant P. americana. Accession numbers: the raw data of P. americana transcriptome presented in this study are openly available in NCBI SRA database, under the BioProject of PRJNA649785. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s13205-021-02865-x.