Application of Deep Learning in Clinical Settings for Detecting and Classifying Malaria Parasites in Thin Blood Smears

Background

Scarcity of annotated image data sets of thin blood smears makes expert-level differentiation among Plasmodium species challenging. Here, we aimed to establish a deep learning algorithm for identifying and classifying malaria parasites in thin blood smears and evaluate its performance and clinical prospect.

Methods

You Only Look Once v7 was used as the backbone network for training the artificial intelligence algorithm model. The training, validation, and test sets for each malaria parasite category were randomly selected. A comprehensive analysis was performed on 12 708 thin blood smear images of various infective stages of 12 546 malaria parasites, including P falciparum, P vivax, P malariae, P ovale, P knowlesi, and P cynomolgi. Peripheral blood samples were obtained from 380 patients diagnosed with malaria. Additionally, blood samples from monkeys diagnosed with malaria were used to analyze P cynomolgi. The accuracy for detecting Plasmodium-infected blood cells was assessed through various evaluation metrics.

Results

The total time to identify 1116 malaria parasites was 13 seconds, with an average analysis time of 0.01 seconds for each parasite in the test set. The average precision was 0.902, with a recall and precision of infected erythrocytes of 96.0% and 94.9%, respectively. Sensitivity and specificity exceeded 96.8% and 99.3%, with an area under the receiver operating characteristic curve >0.999. The highest sensitivity (97.8%) and specificity (99.8%) were observed for trophozoites and merozoites.

Conclusions

The algorithm can help facilitate the clinical and morphologic examination of malaria parasites.

Theoretical Insights into the Solubility Polarity Switch of Metal-Organic Nanoclusters for Nanoscale Patterning

Metal-organic nanoclusters(MOCs) are being increasingly used as prospective photoresist candidates for advanced nanoscale lithography technologies. However, insight into the irradiation-induced solubility switching process remains unclear. Hereby, the theoretical study employing density functional theory (DFT) calculations of the alkene-containing zirconium oxide MOC photoresists is reported, which is rationally synthesized accordingly, to disclose the mechanism of the nanoscale patterning driven by the switch of solubility from the acid-catalyzed or electron-triggered ligand dissociation. By evaluating the dependence of MOCs' imaging process on photoacid, lithographies of photoresists with and without photoacid generators after exposure to ultraviolet (UV), electron beam, and soft X-ray, it is revealed that photoacid is essential in UV lithography, but it demonstrates little effect on exposure dose in high-energy lithography. Furthermore, theoretical studies using DFT simulations to investigate the plausible photoacid-catalyzed, electron-triggered dissociation, and accompanying radical reaction are performed, and a mechanism is demonstrated that the nanoscale patterning of this type of MOCs is driven by the solubility switch resulting from dissociation-induced strong electrostatic interaction and low-energy barrier radical polymerization with other species. This study can give insights into the chemical mechanisms of patterning, and guide the rational design of photoresists to realize high resolution and high sensitivity.

A universal route to efficient non-linear response via Thomson scattering in linear solids

Non-linear materials are cornerstones of modern optics and electronics. Strong dependence on the intrinsic properties of particular materials, however, inhibits the at-will extension of demanding non-linear effects, especially those second-order ones, to widely adopted centrosymmetric materials (for example, silicon) and technologically important burgeoning spectral domains (for example, terahertz frequencies). Here we introduce a universal route to efficient non-linear responses enabled by exciting non-linear Thomson scattering, a fundamental process in electrodynamics that was known to occur only in relativistic electrons in metamaterial composed of linear materials. Such a mechanism modulates the trajectory of charges, either intrinsically or extrinsically provided in solids, at twice the driving frequency, allowing second-harmonic generation at terahertz frequencies on crystalline silicon with extremely large non-linear susceptibility in our proof-of-concept experiments. By offering a substantially material- and frequency-independent platform, our approach opens new possibilities in the fields of on-demand non-linear optics, terahertz sources, strong field light-solid interactions and integrated photonic circuits.

Canny Algorithm Enabling Precise Offline Line Edge Roughness Acquisition in High-Resolution Lithography

The line edge roughness (LER) is one of the most critical indicators of photoresist imaging performance, and its measurement using a reliable method is of great significance for lithography. However, most studies only investigate photoresist resolution and sensitivity because LER measurements require an expensive and not widely available critical dimension scanning electron microscopy (SEM) technology; thus, the imaging performance of photoresist has not been adequately evaluated. Here, we report an image processing software developed for offline calculation of LER that can analyze lithographic patterns with resolutions up to ∼15 nm. This software can effectively process all graphic files obtained from commonly used SEM machines by utilizing the adjustable double threshold. To realize the effective detection of high-resolution patterns in advanced lithography, we used SEM images generated from extreme ultraviolet and electron beam lithography to develop and validate the software's graphic recognition algorithm. This image processing software can process typical SEM images and produce reliable LER in an efficient and user-friendly manner, constituting a powerful tool for promoting the development of high-performance photoresist materials.