It is still challenging to make accurate diagnosis of biliary atresia (BA) with sonographic gallbladder images particularly in rural area without relevant expertise. To help diagnose BA based on sonographic gallbladder images, an ensembled deep learning model is developed. The model yields a patient-level sensitivity 93.1% and specificity 93.9% [with areas under the receiver operating characteristic curve of 0.956 (95% confidence interval: 0.928-0.977)] on the multi-center external validation dataset, superior to that of human experts. With the help of the model, the performances of human experts with various levels are improved. Moreover, the diagnosis based on smartphone photos of sonographic gallbladder images through a smartphone app and based on video sequences by the model still yields expert-level performances. The ensembled deep learning model in this study provides a solution to help radiologists improve the diagnosis of BA in various clinical application scenarios, particularly in rural and undeveloped regions with limited expertise.

The differentiation between benign and malignant breast tumors extends beyond morphological structures to encompass functional alterations within the nodules. The combination of photoacoustic (PA) imaging and radiomics unveils functional insights and intricate details that are imperceptible to the naked eye. This study aims to assess the efficacy of PA imaging in breast cancer radiomics, focusing on the impact of peritumoral region size on radiomic model accuracy. From January 2022 to November 2023, data were collected from 358 patients with breast nodules, diagnosed via PA/US examination and classified as BI-RADS 3–5. The study used the largest lesion dimension in PA images to define the region of interest, expanded by 2 mm, 5 mm, and 8 mm, for extracting radiomic features. Techniques from statistics and machine learning were applied for feature selection, and logistic regression classifiers were used to build radiomic models. These models integrated both intratumoral and peritumoral data, with logistic regressions identifying key predictive features. The developed nomogram, combining 5 mm peritumoral data with intratumoral and clinical features, showed superior diagnostic performance, achieving an AUC of 0.950 in the training cohort and 0.899 in validation. This model outperformed those based solely on clinical features or other radiomic methods, with the 5 mm peritumoral region proving most effective in identifying malignant nodules. This research demonstrates the significant potential of PA imaging in breast cancer radiomics, especially the advantage of integrating 5 mm peritumoral with intratumoral features. This approach not only surpasses models based on clinical data but also underscores the importance of comprehensive radiomic analysis in accurately characterizing breast nodules.

To explore an effective noninvasive tool for monitoring liver fibrosis of children with biliary atresia (BA) is important but evidences are limited. This study is to investigate the predictive accuracy of supersonic shearwave elastography (SSWE) in liver fibrosis for postoperative patients with BA and to compare it with aspartate aminotransferase to platelet ratio index (APRI) and fibrosis-4 (FIB-4). 24 patients with BA received SSWE and laboratory tests before scheduled for liver biopsy. Spearman rank coefficient and receiver operating characteristic (ROC) were used to analyze data. Metavir scores were F0 in 3, F1 in 2, F2 in 4, F3 in 7 and F4 in 8 patients. FIB-4 failed to correlate with fibrosis stage. The areas under the ROC curves of SSWE, APRI and their combination were 0.79, 0.65 and 0.78 for significant fibrosis, 0.81, 0.64 and 0.76 for advanced fibrosis, 0.82, 0.56 and 0.84 for cirrhosis. SSWE values at biopsy was correlated with platelet count (r = −0.426,P = 0.038), serum albumin (r = −0.670,P < 0.001), total bilirubin (r = 0.419,P = 0.041) and direct bilirubin levels (r = 0.518,P = 0.010) measured at 6 months after liver biopsy. Our results indicate that SSWE is a more promising tool to assess liver fibrosis than APRI and FIB-4 in children with BA.

Chen H, et al. Korean J Radiol. 2021 Jun;22(6):959-969. https://doi.org/10.3348/kjr.2020.0885

Background and Purpose: Sonodynamic therapy (SDT) has been widely used for the noninvasive treatment of solid tumors, but the hypoxic tumor microenvironment limits its therapeutic effect. The current methods of reoxygenation to enhance SDT have limitations, prompting reconsideration of the design of therapeutic approaches. Here, we developed a tumor microenvironment-responsive nanoplatform by reducing oxygen consumption to overcome hypoxia-induced resistance to cancer therapy. Methods: A pH-responsive drug-loaded liposome (MI-PEOz-lip) was prepared and used to reduce oxygen consumption, attenuating hypoxia-induced resistance to SDT and thereby improving therapeutic efficiency. Photoacoustic imaging (PAI) and fluorescence imaging (FI) of MI-PEOz-lip were evaluated in vitro and in breast xenograft tumor models. The pH-sensitive functionality of MI-PEOz-lip was applied for pH-triggered cargo release, and its capacity was evaluated. The MI-PEOz-lip-mediated SDT effect was compared with other treatments in vivo. Results: MI-PEOz-lip was demonstrated to specifically accumulate in tumors. Metformin molecules in liposomes selectively accumulate in tumors by pH-responsive drug release to inhibit the mitochondrial respiratory chain while releasing IR780 to the tumor area. These pH-responsive liposomes demonstrated PAI and FI imaging capabilities in vitro and in vivo, providing potential for treatment guidance and monitoring. In particular, the prepared MI-PEOz-lip combined with ultrasound irradiation effectively inhibited breast tumors by producing toxic reactive singlet oxygen species (ROS), while the introduction of metformin inhibited mitochondrial respiration and reduced tumor oxygen consumption, resulting in excellent sonodynamic therapy performance compared with other treatments. Conclusion: In this study, we present a novel strategy to achieve high therapeutic efficacy of SDT by the rational design of multifunctional nanoplatforms. This work provides a new strategy that can solve the current problems of inefficient oxygen delivery strategies and weaken resistance to various oxygen-dependent therapies.