Automatic Lenke classification of adolescent idiopathic scoliosis with deep learning.

Purpose: The Lenke classification system is widely utilized as the preoperative evaluation protocol for adolescent idiopathic scoliosis (AIS). However, manual measurement is susceptible to observer-induced variability, which consequently impacts the evaluation of progression. The goal of this investigation was to develop an automated Lenke classification system utilizing innovative deep learning algorithms. Methods: Using the database from the First Affiliated Hospital of Sun Yat-sen University, the whole spinal x-rays images were retrospectively collected. Specifically, images collection was divided into AIS and control group. The control group consisted of individuals who underwent routine health checks and did not have scoliosis. Afterwards, relative features of all images were annotated. Deep learning was implemented through the utilization of the key-point based detection method to realize the vertebral detection, and Cobb angle measurement and scoliosis classification were performed based on relevant standards. Besides, the segmentation method was employed to achieve the recognition of lumbar vertebral pedicle to determine the type of lumbar spine modifier. Finally, the model performance was further quantitatively analyzed. Results: In the study, a total of 2082 spinal x-ray images were collected from 407 AIS patients and 227 individuals in the control group. The model for vertebral detection achieved an F1-score of 0.809 for curve type evaluation and an F1-score of 0.901 for thoracic sagittal profile. The intraclass correlation efficient (ICC) of the Cobb angle measurement was 0.925. In the analysis of performance for vertebra pedicle segmentation model, the F1-score of lumbar modification profile was 0.942, the intersection over union (IOU) of the target pixels was 0.827, and the Hausdorff distance (HD) was 6.565 ± 2.583 mm. Specifically, the F1-score for ultimate Lenke type classifier was 0.885. Conclusions: This study has constructed an automated Lenke classification system by employing the deep learning networks to achieve the recognition pattern and feature extraction. Our models require further validation in additional cases in the future.

Procedurally Targeted Delivery of Antitumor Drugs Using FAPα-Responsive TPGS Dimer-Based Flower-like Polymeric Micelles.

To overcome the intestinal epithelium barrier and achieve a better antitumor effect, the procedurally targeting flower-like nanomicelles for oral delivery of antitumor drugs were designed based on FAPα-responsive TPGS1000 dimer (TPGS-Gly-Pro-TPGS) and L-carnitine linked poly(2-ethyl-2-oxazoline)-b-poly(D, l-lactide) (Car-PEOz-b-PLA). As expected, compared with unmodified polymeric micelles (TT-PMs) composed of TPGS-Gly-Pro-TPGS, L-carnitine conjugated polymeric micelles (CTT-PMs) formed from both TPGS-Gly-Pro-TPGS and Car-PEOz-b-PLA with favorable stability in simulated gastrointestinal fluid and FAPα-dependent release capability exhibited remarkably enhanced cellular uptake and transmembrane transport through OCTN2 mediation confirmed by fluorescence immunoassay, which was intuitively evidenced by stronger fluorescence within epithelial cells, and the basal side of small intestinal epithelium of mice being given intragastric administration of DiI-labeled micelles. The transport of CTT-PMs across the intestinal epithelium in an intact form was mediated by clathrin along the intracellular transport pathway of endosome-lysosome-ER-Golgi apparatus. Furthermore, both the increased uptake by FAPα-positive U87MG cells and unchangeable uptake by FAPα-negative C6 cells for coumarin-6 (C-6)/CTT-PMs compared with C-6/TT-PMs evidenced the targeting ability of CTT-PMs to FAPα-positive tumor cells. Both OCTN2-mediation and FAPα-responsiveness were beneficial for polymeric micelles to improve the delivery and therapeutic efficiency of antitumor agents, which was further supported by the remarkable enhancement in in vivo antitumor efficacy via promoting apoptosis of tumor cells for paclitaxel (PTX)-loaded CTT-PMs (PTX/CTT-PMs) with low toxicity compared with PTX/TT-PMs. Our findings offered an alternative design strategy for procedurally targeted delivery of chemotherapeutics by an oral route.

PCCT: Progressive Class-Center Triplet Loss for Imbalanced Medical Image Classification.

Imbalanced training data in medical image diagnosis is a significant challenge for diagnosing rare diseases. For this purpose, we propose a novel two-stage Progressive Class-Center Triplet (PCCT) framework to overcome the class imbalance issue. In the first stage, PCCT designs a class-balanced triplet loss to coarsely separate distributions of different classes. Triplets are sampled equally for each class at each training iteration, which alleviates the imbalanced data issue and lays solid foundation for the successive stage. In the second stage, PCCT further designs a class-center involved triplet strategy to enable a more compact distribution for each class. The positive and negative samples in each triplet are replaced by their corresponding class centers, which prompts compact class representations and benefits training stability. The idea of class-center involved loss can be extended to the pair-wise ranking loss and the quadruplet loss, which demonstrates the generalization of the proposed framework. Extensive experiments support that the PCCT framework works effectively for medical image classification with imbalanced training images. On four challenging class-imbalanced datasets (two skin datasets Skin7 and Skin 198, one chest X-ray dataset ChestXray-COVID, and one eye dataset Kaggle EyePACs), the proposed approach respectively obtains the mean F1 score 86.20, 65.20, 91.32, and 87.18 over all classes and 81.40, 63.87, 82.62, and 79.09 for rare classes, achieving state-of-the-art performance and outperforming the widely used methods for the class imbalance issue.

Enhanced Oral Absorption and Liver Distribution of Polymeric Nanoparticles through Traveling the Enterohepatic Circulation Pathways of Bile Acid.

The intestinal epithelium is known to be a main hindrance to oral delivery of nanoparticles. Even though surface ligand modification can enhance cellular uptake of nanoparticles, the "easy entry and hard across" was frequently observed for many active targeting nanoparticles. Here, we fabricated polymeric nanoparticles relayed by bile acid transporters with monomethoxy poly(ethylene glycol)-poly(D,l-lactide) and deoxycholic acid-conjugated poly(2-ethyl-2-oxazoline)-poly(D,l-lactide) based on structural characteristics of intestine epithelium and the absorption characteristics of endogenous substances. As anticipated, deoxycholic acid-modified polymeric nanoparticles featuring good stability in simulated gastrointestinal fluid could notably promote the internalization of their payload by Caco-2 cells through mediation of apical sodium-dependent bile acid transporter (ASBT) and transmembrane transport of the nanoparticles across Caco-2 cell monolayers via relay-guide of ASBT, ileal bile acid-binding protein, and the heteromeric organic solute transporter (OSTα-OSTß) along with multidrug resistance-associated protein 3 (MRP3) evidenced by competitive inhibition and fluorescence immunoassay, which was further visually confirmed by the stronger fluorescence from C6-labeled nanoparticles inside enterocytes and the basal side of the intestinal epithelium of mice. The transcellular transport of deoxycholic acid-modified nanoparticles in an intact form was mediated by caveolin/lipid rafts and clathrin with intracellular trafficking trace of endosome-lysosome-ER-Golgi apparatus and bile acid transport route. Furthermore, the increased uptake by HepG2 cells compared with unmodified nanoparticles evidenced the target ability of deoxycholic acid-modified nanoparticles to the liver, which was further supported by ex vivo imaging of excised major organs of mice. Thus, this study provided a feasible and potential strategy to further enhance transepithelial transport efficiency and liver-targeted ability of nanoparticles by means of the specific enterohepatic circulation pathways of bile acid.