Ultranarrow Deep-Blue Luminescence of Perovskite Nanocrystals by A-Site Cation Control.

Metal-halide perovskite nanocrystals (NCs) are one of the most promising emitters for the application of display and nanolight sources. The full width at half-maximum (FWHM) of photoluminescence (PL) emission is essential for color purity, which however remains a difficulty to further reduce the FWHM of the perovskite NCs at room temperature. Here, we show the quasi-sphere perovskite NCs with narrow PL emission at a deep-blue wavelength of ∼430 nm; its PL FWHM reaches ∼11 nm at room temperature, owing to the monodispersion in size distribution as well as the symmetric quasi-sphere morphology of NCs releasing the fine structure splitting-induced inhomogeneous broadening. Through regulating A cations with respect to the ratio of FA (or MA)-to-Cs and Cs-to-Pb, the PL emission of the NCs could be tuned from ∼505 to ∼430 nm combined with varied morphologies from large cube to small quasi-sphere. Such spectroscopic and morphological discrepancies are supposed to be attributed to the different crystalline kinetics that is strongly dependent on the synthetic condition. To be specific, in the case of increasing FA (or MA)-to-Cs, the growth rate of CsPbBr3 and FAPbBr3 (or MAPbBr3) perovskites is determined by the reactivity of transient species, while in the case of decreasing the Cs-to-Pb ratio, the growth rate of perovskites is slowed down by the serious reduction of Cs+ in the precursor. This study provides an effective strategy to adjust the emission across from green to deep-blue color and promotes the perovskite NCs with a narrow FWHM, and tunable PL emission facilitates in application of optoelectronic devices.

Automatic detection of small bowel lesions with different bleeding risks based on deep learning models.

BACKGROUND: Deep learning provides an efficient automatic image recognition method for small bowel (SB) capsule endoscopy (CE) that can assist physicians in diagnosis. However, the existing deep learning models present some unresolved challenges. AIM: To propose a novel and effective classification and detection model to automatically identify various SB lesions and their bleeding risks, and label the lesions accurately so as to enhance the diagnostic efficiency of physicians and the ability to identify high-risk bleeding groups. METHODS: The proposed model represents a two-stage method that combined image classification with object detection. First, we utilized the improved ResNet-50 classification model to classify endoscopic images into SB lesion images, normal SB mucosa images, and invalid images. Then, the improved YOLO-V5 detection model was utilized to detect the type of lesion and its risk of bleeding, and the location of the lesion was marked. We constructed training and testing sets and compared model-assisted reading with physician reading. RESULTS: The accuracy of the model constructed in this study reached 98.96%, which was higher than the accuracy of other systems using only a single module. The sensitivity, specificity, and accuracy of the model-assisted reading detection of all images were 99.17%, 99.92%, and 99.86%, which were significantly higher than those of the endoscopists' diagnoses. The image processing time of the model was 48 ms/image, and the image processing time of the physicians was 0.40 ± 0.24 s/image (P < 0.001). CONCLUSION: The deep learning model of image classification combined with object detection exhibits a satisfactory diagnostic effect on a variety of SB lesions and their bleeding risks in CE images, which enhances the diagnostic efficiency of physicians and improves the ability of physicians to identify high-risk bleeding groups.