Novel 3D-based deep learning for classification of acute exacerbation of idiopathic pulmonary fibrosis using high-resolution CT.

PURPOSE: Acute exacerbation of idiopathic pulmonary fibrosis (AE-IPF) is the primary cause of death in patients with IPF, characterised by diffuse, bilateral ground-glass opacification on high-resolution CT (HRCT). This study proposes a three-dimensional (3D)-based deep learning algorithm for classifying AE-IPF using HRCT images. MATERIALS AND METHODS: A novel 3D-based deep learning algorithm, SlowFast, was developed by applying a database of 306 HRCT scans obtained from two centres. The scans were divided into four separate subsets (training set, n=105; internal validation set, n=26; temporal test set 1, n=79; and geographical test set 2, n=96). The final training data set consisted of 1050 samples with 33 600 images for algorithm training. Algorithm performance was evaluated using accuracy, sensitivity, specificity, positive predictive value, negative predictive value, receiver operating characteristic (ROC) curve and weighted κ coefficient. RESULTS: The accuracy of the algorithm in classifying AE-IPF on the test sets 1 and 2 was 93.9% and 86.5%, respectively. Interobserver agreements between the algorithm and the majority opinion of the radiologists were good (κw=0.90 for test set 1 and κw=0.73 for test set 2, respectively). The ROC accuracy of the algorithm for classifying AE-IPF on the test sets 1 and 2 was 0.96 and 0.92, respectively. The algorithm performance was superior to visual analysis in accurately diagnosing radiological findings. Furthermore, the algorithm's categorisation was a significant predictor of IPF progression. CONCLUSIONS: The deep learning algorithm provides high auxiliary diagnostic efficiency in patients with AE-IPF and may serve as a useful clinical aid for diagnosis.

The gene expression of CALD1, CDH2, and POSTN in fibroblast are related to idiopathic pulmonary fibrosis.

Introduction: Idiopathic pulmonary fibrosis (IPF) is characterized by progressive lung dysfunction due to excessive collagen production and tissue scarring. Despite recent advancements, the molecular mechanisms remain unclear. Methods: RNA sequencing identified 475 differentially expressed genes (DEGs) in the TGF-ß1-induced primary lung fibrosis model. Gene expression chips GSE101286 and GSE110147 from NCBI gene expression omnibus (GEO) database were analyzed using GEO2R, revealing 94 DEGs in IPF lung tissue samples. The gene ontology (GO) and pathway enrichment, Protein-protein interaction (PPI) network construction, and Maximal Clique Centrality (MCC) scoring were performed. Experimental validation included RT-qPCR, Immunohistochemistry (IHC), and Western Blot, with siRNA used for gene knockdown. A co-expression network was constructed by GeneMANIA. Results: GO enrichment highlighted significant enrichment of DEGs in TGF-ß cellular response, connective tissue development, extracellular matrix components, and signaling pathways such as the AGE-RAGE signaling pathway and ECM-receptor interaction. PPI network analysis identified hub genes, including FN1, COL1A1, POSTN, KIF11, and ECT2. CALD1 (Caldesmon 1), CDH2 (Cadherin 2), and POSTN (Periostin) were identified as dysregulated hub genes in both the RNA sequencing and GEO datasets. Validation experiments confirmed the upregulation of CALD1, CDH2, and POSTN in TGF-ß1-treated fibroblasts and IPF lung tissue samples. IHC experiments probed tissue-level expression patterns of these three molecules. Knockdown of CALD1, CDH2, and POSTN attenuated the expression of fibrotic markers (collagen I and α-SMA) in response to TGF-ß1 stimulation in primary fibroblasts. Co-expression analysis revealed interactions between hub genes and predicted genes involved in actin cytoskeleton regulation and cell-cell junction organization. Conclusions: CALD1, CDH2, and POSTN, identified as potential contributors to pulmonary fibrosis, present promising therapeutic targets for IPF patients.

Animal models of acute exacerbation of pulmonary fibrosis.

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive scarring interstitial lung disease with an unknown cause. Some patients may experience acute exacerbations (AE), which result in severe lung damage visible on imaging or through examination of tissue samples, often leading to high mortality rates. However, the etiology and pathogenesis of AE-IPF remain unclear. AE-IPF patients exhibit diffuse lung damage, apoptosis of type II alveolar epithelial cells, and an excessive inflammatory response. Establishing a reliable animal model of AE is critical for investigating the pathogenesis. Recent studies have reported a variety of animal models for AE-IPF, each with its own advantages and disadvantages. These models are usually established in mice with bleomycin-induced pulmonary fibrosis, using viruses, bacteria, small peptides, or specific drugs. In this review, we present an overview of different AE models, hoping to provide a useful resource for exploring the mechanisms and targeted therapies for AE-IPF.

Correction: Ou et al. Characteristic Microbiomes Correlate with Polyphosphate Accumulation of Marine Sponges in South China Sea Areas. Microorganisms 2020, 8, 63.

The authors wish to make the following corrections to this paper [...].

Characteristic Microbiomes Correlate with Polyphosphate Accumulation of Marine Sponges in South China Sea Areas.

Some sponges have been shown to accumulate abundant phosphorus in the form of polyphosphate (polyP) granules even in waters where phosphorus is present at low concentrations. But the polyP accumulation occurring in sponges and their symbiotic bacteria have been little studied. The amounts of polyP exhibited significant differences in twelve sponges from marine environments with high or low dissolved inorganic phosphorus (DIP) concentrations which were quantified by spectral analysis, even though in the same sponge genus, e.g., Mycale sp. or Callyspongia sp. PolyP enrichment rates of sponges in oligotrophic environments were far higher than those in eutrophic environments. Massive polyP granules were observed under confocal microscopy in samples from very low DIP environments. The composition of sponge symbiotic microbes was analyzed by high-throughput sequencing and the corresponding polyphosphate kinase (ppk) genes were detected. Sequence analysis revealed that in the low DIP environment, those sponges with higher polyP content and enrichment rates had relatively higher abundances of cyanobacteria. Mantel tests and canonical correspondence analysis (CCA) examined that the polyP enrichment rate was most strongly correlated with the structure of microbial communities, including genera Synechococcus, Rhodopirellula, Blastopirellula, and Rubripirellula. About 50% of ppk genes obtained from the total DNA of sponge holobionts, had above 80% amino acid sequence similarities to those sequences from Synechococcus. In general, it suggested that sponges employed differentiated strategies towards the use of phosphorus in different nutrient environments and the symbiotic Synechococcus could play a key role in accumulating polyP.