Corrigendum: Ultrasound-based radiomics analysis for differentiating benign and malignant breast lesions: from static images to CEUS video analysis.

[This corrects the article DOI: 10.3389/fonc.2022.951973.].

Ultrasound-based radiomics analysis for differentiating benign and malignant breast lesions: From static images to CEUS video analysis.

Background: Continuous contrast-enhanced ultrasound (CEUS) video is a challenging direction for radiomics research. We aimed to evaluate machine learning (ML) approaches with radiomics combined with the XGBoost model and a convolutional neural network (CNN) for discriminating between benign and malignant lesions in CEUS videos with a duration of more than 1 min. Methods: We gathered breast CEUS videos of 109 benign and 81 malignant tumors from two centers. Radiomics combined with the XGBoost model and a CNN was used to classify the breast lesions on the CEUS videos. The lesions were manually segmented by one radiologist. Radiomics combined with the XGBoost model was conducted with a variety of data sampling methods. The CNN used pretrained 3D residual network (ResNet) models with 18, 34, 50, and 101 layers. The machine interpretations were compared with prospective interpretations by two radiologists. Breast biopsies or pathological examinations were used as the reference standard. Areas under the receiver operating curves (AUCs) were used to compare the diagnostic performance of the models. Results: The CNN model achieved the best AUC of 0.84 on the test cohort with the 3D-ResNet-50 model. The radiomics model obtained AUCs between 0.65 and 0.75. Radiologists 1 and 2 had AUCs of 0.75 and 0.70, respectively. Conclusions: The 3D-ResNet-50 model was superior to the radiomics combined with the XGBoost model in classifying enhanced lesions as benign or malignant on CEUS videos. The CNN model was superior to the radiologists, and the radiomics model performance was close to the performance of the radiologists.

Protective effect of dexmedetomidine in cecal ligation perforation-induced acute lung injury through HMGB1/RAGE pathway regulation and pyroptosis activation.

Dexmedetomidine (DEX) has been reported to attenuate cecal ligation perforation (CLP)-stimulated acute lung injury (ALI) by downregulating HMGB1 and RAGE. This study aimed to further investigate the specific mechanisms of RAGE and its potential-related mechanisms of DEX on ALI models in vitro and in vivo. The in vitro and in vivo ALI models were established by lipopolysaccharide treatment in MLE-12 cells and CLP in mice, respectively. The effect of DEX on pathological alteration was investigated by HE staining. Thereafter, the myeloperoxidase (MPO) activity and inflammatory cytokine levels were respectively detected to assess the lung injury of mice using commercial kits. The expression levels of HMGB1, RAGE, NF-κB, and pyroptosis-related molecules were detected by RT-qPCR and Western blot. HE staining showed that lung injury, increased inflammatory cell infiltration, and lung permeability was found in the ALI mice, and DEX treatment significantly attenuated lung tissue damage induced by CLP. The MPO activity and inflammatory cytokines (TNF-α, IL-1ß, and NLRP3) levels were also significantly reduced after DEX treatment compared with those in the ALI mice. Moreover, DEX activated the HMGB1/RAGE/NF-κB pathway and upregulated the pyroptosis-related proteins. However, the protective DEX effect was impaired by RAGE overexpression in ALI mice and MLE-12 cells. Additionally, DEX treatment significantly suppressed HMGB1 translocation from the nucleus region to the cytoplasm, and this effect was reversed by RAGE overexpression. These findings suggested that DEX may be a useful ALI treatment, and the protective effects on ALI mice may be through the inhibition of HMGB1/RAGE/NF-κB pathway and cell pyroptosis.

Stevens-Johnson syndrome/toxic epidermal necrolysis successfully treated with Chinese herbal medicine Pi-Yan-Ning: A case report.

Stevens-Johnson syndrome/toxic epidermal necrolysis (SJS/TEN) is a rare adverse cutaneous reaction with a low incidence and high mortality. Despite posing a serious threat to patients' health and lives, there is no high-quality evidence for a standard treatment regimen. Here we report the case of a 62-year-old man with stage IV pancreatic cancer who experienced immunotherapy-induced SJS/TEN. After consensus-based regular treatments at a local hospital, his symptoms became worse. Thus, he consented to receive Chinese herbal medicine (CHM) therapy. The affected parts of the patient were treated with the CHM Pi-Yan-Ning which was applied externally for 20 min twice a day. After 7 days of treatment, the dead skin began peeling away from the former lesions that had covered his hands, feet, and lips, indicating that skin had regenerated. After 12 days of treatment, the patient's skin was completely recovered. In this case, SJS/TEN was successfully treated with Pi-Yan-Ning, suggesting that there might be tremendous potential for the use of Pi-Yan-Ning in the treatment of severe skin reactions to drug treatments. Further basic investigations and clinical trials to explore the mechanism and efficacy are needed.

Key miRNAs associated with memory and learning disorder upon exposure to sevoflurane determined by RNA sequencing.

The study aimed to identify differentially expressed microRNAs (miRNAs/miRs) and explore the mechanisms governing impaired memory and learning ability in developing brains exposed to sevoflurane. A total of six 7­day­old male ICR mice were randomly assigned into the sevoflurane anesthesia group (treated with 2.4% sevoflurane) or control group (treated with normal saline solution at the same dose). After 14 days, the mice were subjected to a Morris water maze experiment. Then, the animals were sacrificed and hippocampus tissues were isolated. RNAs in hippocampus tissues were sequenced and the differential miRNA expression profiles were identified by a bioinformatics approach. The learning and memory function of mice were significantly affected by sevoflurane exposure. A total of 18 miRNAs were found to be significantly affected by sevoflurane administration. Their target genes clustered into different functional groups, such as 'dephosphorylation', 'vesicle localization' and the 'Wnt signaling pathway'. miR­101b­3p was closely related with 'chromatin binding' and 'protein serine/threonine kinase activity'. The most represented pathways for miRNAs included 'neuroactive ligand­receptor interaction' (miR­1187), 'long­term depression' (miR­425­5p), 'FoxO signaling pathway' (miR­425­5p) and the 'neurotrophin signaling pathway' (miR­467a­3p). miR­467a­3p (degree=89), miR­101b­3p (degree=59), and miR­1187 (degree=51) were the hub nodes in the miRNA regulatory network. The Wnt signaling pathway, miR­467a­3p, miR­1187 and miR­101b­3p may be therapeutic targets for preventing cognitive impairments induced by sevoflurane.