Integrating artificial intelligence and wing geometric morphometry to automate mosquito classification.

Mosquitoes (Diptera: Culicidae) comprise over 3500 global species, primarily in tropical regions, where the females act as disease vectors. Thus, identifying medically significant species is vital. In this context, Wing Geometric Morphometry (WGM) emerges as a precise and accessible method, excelling in species differentiation through mathematical approaches. Computational technologies and Artificial Intelligence (AI) promise to overcome WGM challenges, supporting mosquito identification. AI explores computers' thinking capacity, originating in the 1950s. Machine Learning (ML) arose in the 1980s as a subfield of AI, and deep Learning (DL) characterizes ML's subcategory, featuring hierarchical data processing layers. DL relies on data volume and layer adjustments. Over the past decade, AI demonstrated potential in mosquito identification. Various studies employed optical sensors, and Convolutional Neural Networks (CNNs) for mosquito identification, achieving average accuracy rates between 84 % and 93 %. Furthermore, larval Aedes identification reached accuracy rates of 92 % to 94 % using CNNs. DL models such as ResNet50 and VGG16 achieved up to 95 % accuracy in mosquito identification. Applying CNNs to georeference mosquito photos showed promising results. AI algorithms automated landmark detection in various insects' wings with repeatability rates exceeding 90 %. Companies have developed wing landmark detection algorithms, marking significant advancements in the field. In this review, we discuss how AI and WGM are being combined to identify mosquito species, offering benefits in monitoring and controlling mosquito populations.

Automatic detection of the parasite Trypanosoma cruzi in blood smears using a machine learning approach applied to mobile phone images.

Chagas disease is a life-threatening illness caused by the parasite Trypanosoma cruzi. The diagnosis of the acute form of the disease is performed by trained microscopists who detect parasites in blood smear samples. Since this method requires a dedicated high-resolution camera system attached to the microscope, the diagnostic method is more expensive and often prohibitive for low-income settings. Here, we present a machine learning approach based on a random forest (RF) algorithm for the detection and counting of T. cruzi trypomastigotes in mobile phone images. We analyzed micrographs of blood smear samples that were acquired using a mobile device camera capable of capturing images in a resolution of 12 megapixels. We extracted a set of features that describe morphometric parameters (geometry and curvature), as well as color, and texture measurements of 1,314 parasites. The features were divided into train and test sets (4:1) and classified using the RF algorithm. The values of precision, sensitivity, and area under the receiver operating characteristic (ROC) curve of the proposed method were 87.6%, 90.5%, and 0.942, respectively. Automating image analysis acquired with a mobile device is a viable alternative for reducing costs and gaining efficiency in the use of the optical microscope.

Prognostic value of integrin αV expression and localization pattern in invasive breast carcinomas.

Invasion of surrounding stroma is an early event in breast cancer metastatic progression, and involves loss of cell polarity, loss of myoepithelial layer, epithelial-mesenchymal transition (EMT) and remodeling of the extracellular matrix (ECM). Integrins are transmembrane receptors responsible for cell-ECM binding, which triggers signals that regulate many aspects of cell behavior and fate. Changes in the expression, localization and pairing of integrins contribute for abnormal responses found in transformed epithelia. We analyzed 345 human breast cancer samples in tissue microarrays (TMA) from cases diagnosed with invasive breast carcinoma to assess the expression and localization pattern of integrin αV and correlation with clinical parameters. Patients with lower levels of integrin αV staining showed reduced cancer specific survival. A subset of cases presented a peripheral staining of integrin αV surrounding tumor cell clusters, possibly matching the remaining myoepithelial layer. Indeed, the majority of ductal carcinoma in situ (DCIS) components found in the TMA presented integrin αV at their periphery, whereas this pattern was mostly lost in invasive components, even in the same sample. The lack of peripheral integrin αV correlated with decreased cancer specific survival. In addition, we observed that the presence of integrin αV in the stroma was an indicative of poor survival and metastatic disease. Consistently, by interrogating publicly available datasets we found that, although patients with higher mRNA levels of integrin αV had increased risk of developing metastasis, high co-expression of integrin αV and a myoepithelial cell marker (MYH11) mRNA levels correlated with better clinical outcomes. Finally, a 3D cell culture model of non-malignant and malignant cells reproduced the integrin αV pattern seen in patient samples. Taken together, our data indicate that both the expression levels of integrin αV and its tissue localization in primary tumors have prognostic value, and thus, could be used to help predict patients at higher risk of developing metastasis.

Alkaloids extracted from Pterogyne nitens induce apoptosis in malignant breast cell line.

In the present study, two alkaloids isolated from Pterogyne nitens, a plant native to Brazil, have been shown to induce apoptosis in human breast cancer cells. These compounds, pterogynine (PGN) and pterogynidine (PGD), were tested for their effect on a human infiltrating ductal carcinoma cell line (ZR-7531). The cell line was treated with each alkaloid at several concentrations. Time-dependence (with or without recuperation time) and concentration-dependence (in the range 0.25-10 mM) were investigated in cytotoxicity and apoptosis assays. The annexin assay indicated an apparently higher percentage of death by necrosis of malignant cells after 24 h exposure to both P. nitens extracts than the Hoechst assay. Thus, our results in the two tests demonstrated that the Hoechst assay can discriminate between late apoptotic cells and necrosis, whereas the flow cytometry-based annexin V assay cannot. We concluded that PGN and PGD have effective antineoplastic activity against human breast cancer cells in vitro, by inducing programmed cell death.

Stochastic model of contact inhibition and the proliferation of melanoma in situ.

Contact inhibition is a central feature orchestrating cell proliferation in culture experiments; its loss is associated with malignant transformation and tumorigenesis. We performed a co-culture experiment with human metastatic melanoma cell line (SKMEL- 147) and immortalized keratinocyte cells (HaCaT). After 8 days a spatial pattern was detected, characterized by the formation of clusters of melanoma cells surrounded by keratinocytes constraining their proliferation. In addition, we observed that the proportion of melanoma cells within the total population has increased. To explain our results we propose a spatial stochastic model (following a philosophy of the Widom-Rowlinson model from Statistical Physics and Molecular Chemistry) which considers cell proliferation, death, migration, and cell-to-cell interaction through contact inhibition. Our numerical simulations demonstrate that loss of contact inhibition is a sufficient mechanism, appropriate for an explanation of the increase in the proportion of tumor cells and generation of spatial patterns established in the conducted experiments.