Development of full‐field optical spatial coherence tomography system for automated identification of malaria using the multilevel ensemble classifier

Malaria is a life‐threatening infectious blood disease affecting humans and other animals caused by parasitic protozoans belonging to the Plasmodium type especially in developing countries. The gold standard method for the detection of malaria is through the microscopic method of chemically treated blood smears. We developed an automated optical spatial coherence tomographic system using a machine learning approach for a fast identification of malaria cells. In this study, 28 samples (15 healthy, 13 malaria infected stages of red blood cells) were imaged by the developed system and 13 features were extracted. We designed a multilevel ensemble‐based classifier for the quantitative prediction of different stages of the malaria cells. The proposed classifier was used by repeating k‐fold cross validation dataset and achieve a high‐average accuracy of 97.9% for identifying malaria infected late trophozoite stage of cells. Overall, our proposed system and multilevel ensemble model has a substantial quantifiable potential to detect the different stages of malaria infection without staining or expert.      

Stiffening of Red Blood Cells Induced by Cytoskeleton Disorders: A Joint Theory-Experiment Study

The functions and elasticities of the cell are largely related to the structures of the cytoskeletons underlying the lipid bilayer. Among various cell types, the red blood cell (RBC) possesses a relatively simple cytoskeletal structure. Underneath the membrane, the RBC cytoskeleton takes the form of a two-dimensional triangular network, consisting of nodes of actins (and other proteins) and edges of spectrins. Recent experiments focusing on the malaria-infected RBCs (iRBCs) show that there is a correlation between the elongation of spectrins in the cytoskeletal network and the stiffening of the iRBCs. Here we rationalize the correlation between these two observations by combining the wormlike chain model for single spectrins and the effective medium theory for the network elasticity. We specifically focus on how the disorders in the cytoskeletal network affect its macroscopic elasticity. Analytical and numerical solutions from our model reveal that the stiffness of the membrane increases with increasing end-to-end distances of spectrins, but has a nonmonotonic dependence on the variance of the end-to-end distance distributions. These predictions are verified quantitatively by our atomic force microscopy and micropipette aspiration measurements of iRBCs. The model may, from a molecular level, provide guidelines for future identification of new treatment methods for RBC-related diseases, such as malaria infection.     

Approval policies for modifications to machine learning‐based software as a medical device: A study of bio‐creep

Successful deployment of machine learning algorithms in healthcare requires careful assessments of their performance and safety. To date, the FDA approves locked algorithms prior to marketing and requires future updates to undergo separate premarket reviews. However, this negates a key feature of machine learning—the ability to learn from a growing dataset and improve over time. This paper frames the design of an approval policy, which we refer to as an automatic algorithmic change protocol (aACP), as an online hypothesis testing problem. As this process has obvious analogy with noninferiority testing of new drugs, we investigate how repeated testing and adoption of modifications might lead to gradual deterioration in prediction accuracy, also known as “biocreep” in the drug development literature. We consider simple policies that one might consider but do not necessarily offer any error‐rate guarantees, as well as policies that do provide error‐rate control. For the latter, we define two online error‐rates appropriate for this context: bad approval count (BAC) and bad approval and benchmark ratios (BABR). We control these rates in the simple setting of a constant population and data source using policies aACP‐BAC and aACP‐BABR, which combine alpha‐investing, group‐sequential, and gate‐keeping methods. In simulation studies, bio‐creep regularly occurred when using policies with no error‐rate guarantees, whereas aACP‐BAC and aACP‐BABR controlled the rate of bio‐creep without substantially impacting our ability to approve beneficial modifications.     

Machine learning of large‐scale spatial distributions of wild turkeys with high‐dimensional environmental data

Species distribution modeling often involves high‐dimensional environmental data. Large amounts of data and multicollinearity among covariates impose challenges to statistical models in variable selection for reliable inferences of the effects of environmental factors on the spatial distribution of species. Few studies have evaluated and compared the performance of multiple machine learning (ML) models in handling multicollinearity. Here, we assessed the effectiveness of removal of correlated covariates and regularization to cope with multicollinearity in ML models for habitat suitability. Three machine learning algorithms maximum entropy (MaxEnt), random forests (RFs), and support vector machines (SVMs) were applied to the original data (OD) of 27 landscape variables, reduced data (RD) with 14 highly correlated covariates being removed, and 15 principal components (PC) of the OD accounting for 90% of the original variability. The performance of the three ML models was measured with the area under the curve and continuous Boyce index. We collected 663 nonduplicated presence locations of Eastern wild turkeys (Meleagris gallopavo silvestris) across the state of Mississippi, United States. Of the total locations, 453 locations separated by a distance of ≥2 km were used to train the three ML algorithms on the OD, RD, and PC data, respectively. The remaining 210 locations were used to validate the trained ML models to measure ML performance. Three ML models had excellent performance on the RD and PC data. MaxEnt and SVMs had good performance on the OD data, indicating the adequacy of regularization of the default setting for multicollinearity. Weak learning of RFs through bagging appeared to alleviate multicollinearity and resulted in excellent performance on the OD data. Regularization of ML algorithms may help exploratory studies of the effects of environmental factors on the spatial distribution and habitat suitability of wildlife.     

Competing for blood: the ecology of parasite resource competition in human malaria–helminth co‐infections

Ecological theory suggests that co‐infecting parasite species can interact within hosts directly, via host immunity and/or via resource competition. In mice, competition for red blood cells (RBCs) between malaria and bloodsucking helminths can regulate malaria population dynamics, but the importance of RBC competition in human hosts was unknown. We analysed infection density (i.e. the concentration of parasites in infected hosts), from a 2‐year deworming study of over 4000 human subjects. After accounting for resource‐use differences among parasites, we find evidence of resource competition, priority effects and a competitive hierarchy within co‐infected individuals. For example reducing competition via deworming increased Plasmodium vivax densities 2.8‐fold, and this effect is limited to bloodsucking hookworms. Our ecological, resource‐based perspective sheds new light into decades of conflicting outcomes of malaria–helminth co‐infection studies with significant health and transmission consequences. Beyond blood, investigating within‐human resource competition may bring new insights for improving human health.     

Novel N‐Acyl‐1H‐imidazole‐1‐carbothioamides: Design,Synthesis, Biological and Computational Studies

The present study reports the convenient synthesis, spectroscopic characterization, bio‐assays and computational evaluation of a novel series of N‐acyl‐1H‐imidazole‐1‐carbothioamides. The screened derivatives displayed excellent antioxidant activity, moderate antibacterial and antifungal potential. The screened derivatives were found to be highly biocompatible against hRBCs. Molecular docking ascertained the mechanism and mode of action towards the molecular target delineating that ligands and complexes were stabilized at the active site by electrostatic and hydrophobic forces in accordance to the corresponding experimental results. Docking simulation provided additional information about the possibilities of inhibitory potential of the compounds against RNA. Computational evaluation predicted that N‐acyl‐1H‐imidazole‐1‐carbothioamides 5c and 5g can serve as potential surrogates for hit to lead generation and design of novel antioxidant and antibacterial agents.     

DNABind: A hybrid algorithm for structure‐based prediction of DNA‐binding residues by combining machine learning‐ and template‐based approaches

Accurate prediction of DNA‐binding residues has become a problem of increasing importance in structural bioinformatics. Here, we presented DNABind, a novel hybrid algorithm for identifying these crucial residues by exploiting the complementarity between machine learning‐ and template‐based methods. Our machine learning‐based method was based on the probabilistic combination of a structure‐based and a sequence‐based predictor, both of which were implemented using support vector machines algorithms. The former included our well‐designed structural features, such as solvent accessibility, local geometry, topological features, and relative positions, which can effectively quantify the difference between DNA‐binding and nonbinding residues. The latter combined evolutionary conservation features with three other sequence attributes. Our template‐based method depended on structural alignment and utilized the template structure from known protein–DNA complexes to infer DNA‐binding residues. We showed that the template method had excellent performance when reliable templates were found for the query proteins but tended to be strongly influenced by the template quality as well as the conformational changes upon DNA binding. In contrast, the machine learning approach yielded better performance when high‐quality templates were not available (about 1/3 cases in our dataset) or the query protein was subject to intensive transformation changes upon DNA binding. Our extensive experiments indicated that the hybrid approach can distinctly improve the performance of the individual methods for both bound and unbound structures. DNABind also significantly outperformed the state‐of‐art algorithms by around 10% in terms of Matthews's correlation coefficient. The proposed methodology could also have wide application in various protein functional site annotations. DNABind is freely available at http://mleg.cse.sc.edu/DNABind/ . Proteins 2013; 81:1885–1899. © 2013 Wiley Periodicals, Inc.     

Insights from Machine Learning Techniques for Predicting the Efficiency of Fullerene Derivatives‐Based Ternary Organic Solar Cells at Ternary Blend Design

Ternary organic solar cells (OSCs) have progressed significantly in recent years due to the sufficient photon harvesting of the blend photoactive layer including three absorption‐complementary materials. With the rapid development of highly efficient ternary OSCs in photovoltaics, the precise energy‐level alignment of the three active components within ternary OSC devices should be taken into account. The machine‐learning technique is a computational method that can effectively learn from previous historical data to build predictive models. In this study, a dataset of 124 fullerene derivatives‐based ternary OSCs is manually constructed from a diverse range of literature along with their frontier molecular orbital theory levels, and device structures. Different machine‐learning algorithms are trained based on these electronic parameters to predict photovoltaic efficiency. Thus, the best predictive capability is provided by using the Random Forest approach beyond other machine‐learning algorithms in the dataset. Furthermore, the Random Forest algorithm yields valuable insights into the crucial role of lowest unoccupied molecular orbital energy levels of organic donors in the performance of ternary OSCs. The outcome of this study demonstrates a smart strategy for extracting underlying complex correlations in fullerene derivatives‐based ternary OSCs, thereby accelerating the development of ternary OSCs and related research fields.     

Multimodal analysis of Plasmodium knowlesi‐infected erythrocytes reveals large invaginations,swelling of the host cell,and rheological defects

The simian parasite Plasmodium knowlesi causes severe and fatal malaria infections in humans, but the process of host cell remodelling that underpins the pathology of this zoonotic parasite is only poorly understood. We have used serial block‐face scanning electron microscopy to explore the topography of P. knowlesi‐infected red blood cells (RBCs) at different stages of asexual development. The parasite elaborates large flattened cisternae (Sinton Mulligan's clefts) and tubular vesicles in the host cell cytoplasm, as well as parasitophorous vacuole membrane bulges and blebs, and caveolar structures at the RBC membrane. Large invaginations of host RBC cytoplasm are formed early in development, both from classical cytostomal structures and from larger stabilised pores. Although degradation of haemoglobin is observed in multiple disconnected digestive vacuoles, the persistence of large invaginations during development suggests inefficient consumption of the host cell cytoplasm. The parasite eventually occupies ~40% of the host RBC volume, inducing a 20% increase in volume of the host RBC and an 11% decrease in the surface area to volume ratio, which collectively decreases the ability of the P. knowlesi‐infected RBCs to enter small capillaries of a human erythrocyte microchannel analyser. Ektacytometry reveals a markedly decreased deformability, whereas correlative light microscopy/scanning electron microscopy and python‐based skeleton analysis (Skan) reveal modifications to the surface of infected RBCs that underpin these physical changes. We show that P. knowlesi‐infected RBCs are refractory to treatment with sorbitol lysis but are hypersensitive to hypotonic lysis. The observed physical changes in the host RBCs may underpin the pathology observed in patients infected with P. knowlesi.     

Estimating physical splenic filtration of Plasmodium falciparum‐infected red blood cells in malaria patients

Splenic filtration of Plasmodium falciparum‐infected red blood cells has been hypothesized to influence malaria pathogenesis. We have developed a minimum cylindrical diameter (MCD) filtration model which estimates physical splenic filtration during malaria infection. The key parameter in the model is the MCD, the smallest tube or cylinder that a red blood cell (RBC) can traverse without lysing. The MCD is defined by a relationship between the RBC surface area and volume. In the MCD filtration model, the MCD filtration function represents the probability of a cell becoming physically removed from circulation. This modelling approach was implemented at a field site in Blantyre, Malawi. We analysed peripheral blood samples from 120 study participants in four clinically defined groups (30 subjects each): cerebral malaria, uncomplicated malaria, aparasitaemic coma and healthy controls. We found statistically significant differences in the surface area and volumes of uninfected RBCs when healthy controls were compared with malaria patients. The estimated filtration rates generated by the MCD model corresponded to previous observations in ex vivo spleen experiments and models of red blood cell loss during acute malaria anaemia.There were no differences in the estimated splenic filtration rates between cerebral malaria and uncomplicated malaria patients. The MCD filtration model estimates that at time of admission, one ring‐stage infected RBC is physically filtered by the spleen for each parasite that remains in peripheral circulation. This field study is the first to use microfluidic devices to identify rheological diversity in RBC populations associated with malaria infection and illness in well‐characterized groups of children living in a malaria endemic area.     

Determination of causes of death via spectrochemical analysis of forensic autopsies‐based pulmonary edema fluid samples with deep learning algorithm

This study investigated whether infrared spectroscopy combined with a deep learning algorithm could be a useful tool for determining causes of death by analyzing pulmonary edema fluid from forensic autopsies. A newly designed convolutional neural network‐based deep learning framework, named DeepIR and eight popular machine learning algorithms, were used to construct classifiers. The prediction performances of these classifiers demonstrated that DeepIR outperformed the machine learning algorithms in establishing classifiers to determine the causes of death. Moreover, DeepIR was generally less dependent on preprocessing procedures than were the machine learning algorithms; it provided the validation accuracy with a narrow range from 0.9661 to 0.9856 and the test accuracy ranging from 0.8774 to 0.9167 on the raw pulmonary edema fluid spectral dataset and the nine preprocessing protocol‐based datasets in our study. In conclusion, this study demonstrates that the deep learning‐equipped Fourier transform infrared spectroscopy technique has the potential to be an effective aid for determining causes of death.     

Optofluidic single‐cell absorption flow analyzer for point‐of‐care diagnosis of malaria

In this work, an optofluidic flow analyzer, which can be used to perform malaria diagnosis at the point‐of‐care is demonstrated. The presented technique is based on quantitative optical absorption measurements carried out on a single cell level for a given population of Human Red Blood Cells (RBCs). By measuring the optical absorption of each RBC, the decrease in the Hemoglobin (Hb) concentration in the cytoplasm of the cell due to the invasion of malarial parasite is detected. Cells are assessed on a single cell basis, as they pass through a microfluidic channel. The proposed technique has been implemented with inexpensive off‐the‐shelf components like laser diode, photo‐detector and a micro‐controller. The ability of the optofluidic flow analyzer to asses about 308,049 cells within 3 minutes has been demonstrated. The presented technique is capable of detecting very low parasitemia levels with high sensitivity.

     

Convolutional neural network‐based malaria diagnosis from focus stack of blood smear images acquired using custom‐built slide scanner

The present paper introduces a focus stacking‐based approach for automated quantitative detection of Plasmodium falciparum malaria from blood smear. For the detection, a custom designed convolutional neural network (CNN) operating on focus stack of images is used. The cell counting problem is addressed as the segmentation problem and we propose a 2‐level segmentation strategy. Use of CNN operating on focus stack for the detection of malaria is first of its kind, and it not only improved the detection accuracy (both in terms of sensitivity [97.06%] and specificity [98.50%]) but also favored the processing on cell patches and avoided the need for hand‐engineered features. The slide images are acquired with a custom‐built portable slide scanner made from low‐cost, off‐the‐shelf components and is suitable for point‐of‐care diagnostics. The proposed approach of employing sophisticated algorithmic processing together with inexpensive instrumentation can potentially benefit clinicians to enable malaria diagnosis.      

Inside Cover: A low‐cost,automated parasite diagnostic system via a portable,robotic microscope and deep learning (J. Biophotonics 9/2019)

A low‐cost, automated microscope is combined with machine learning to bring veterinary parasite diagnosis to the point of need. The authors present an inexpensive robotic microscope that automatically focuses, scans, and images a large area McMaster chamber. A deep learning image segmentation pipeline identifies and counts eggs of parasitic worms and single‐celled parasites in goats, dogs, and monkeys, yielding >96% diagnostic accuracy without the need for a trained user. Further details can be found in the article by Yaning Li, Rui Zheng, Yizhen Wu, et al. ( e201800410 ).

     

Conserved regions from Plasmodium falciparum MSP11 specifically interact with host cells and have a potential role during merozoite invasion of red blood cells

Despite significant global efforts, a completely effective vaccine against Plasmodium falciparum, the species responsible for the most serious form of malaria, has not been yet obtained. One of the most promising approaches consists in combining chemically synthesized minimal subunits of parasite proteins involved in host cell invasion, which has led to the identification of peptides with high binding activity (named HABPs) to hepatocyte and red blood cell (RBC) surface receptors in a large number of sporozoite and merozoite proteins, respectively. Among these proteins is the merozoite surface protein 11 (MSP11), which shares important structural and immunological features with the antimalarial vaccine candidates MSP1, MSP3, and MSP6. In this study, 20‐mer‐long synthetic peptides spanning the complete sequence of MSP11 were assessed for their ability to bind specifically to RBCs. Two HABPs with high ability to inhibit invasion of RBCs in vitro were identified (namely HABPs 33595 and 33606). HABP‐RBC bindings were characterized by means of saturation assays and Hill analysis, finding cooperative interactions of high affinity for both HABPs (n_H of 1.5 and 1.2, K_d of 800 and 600 nM for HABPs 33595 and 33606, respectively). The nature of the possible RBC receptors for MSP11 HABPs was studied in binding assays to enzyme‐treated RBCs and cross‐linking assays, finding that both HABPs use mainly a sialic acid‐dependent receptor. An analysis of the immunological, structural and polymorphic characteristics of MSP11 HABPs supports including these peptides in further studies with the aim of designing a fully effective protection‐inducing vaccine against malaria. J. Cell. Biochem. 110: 882–892, 2010. © 2010 Wiley‐Liss, Inc.     

Diagnosis of early gastric cancer based on fluorescence hyperspectral imaging technology combined with partial‐least‐square discriminant analysis and support vector machine

This study investigated the feasibility of using fluorescence hyperspectral imaging technology to diagnose of early‐stage gastric cancer. Fluorescence spectral images of 76 patients who were pathologically diagnosed as non‐atrophic gastritis, premalignant lesions and gastric cancer were collected. Fluorescence spectra at 100‐pixel points were randomly extracted after binarization. Diagnostic models of non‐atrophic gastritis, premalignant lesions and gastric cancer were constructed through partial‐least‐square discriminant analysis (PLS‐DA) and support vector machine (SVM) algorithms. The prediction effects of PLS‐DA and SVM models were compared. Results showed that the average spectra of normal, precancerous and gastric cancer tissues significantly differed at 496, 546, 640 and 670 nm, and regular changes in fluorescence intensity at 546 nm were in the following order: normal > precancerous lesions > gastric cancer. Additionally, the effect of the diagnostic model established by SVM is significantly better than PLS‐DA which accuracy, specificity and sensitivity are above 94%. Experimental results revealed that the fast diagnostic model of early gastric cancer by combining fluorescence hyperspectral imaging technology and improved SVM was effective and feasible, thereby providing an accurate and rapid method for diagnosing early‐stage gastric cancer.