Thermal and oxygen conditions during development cause common rough woodlice (Porcellio scaber) to alter the size of their gas-exchange organs.

Terrestrial isopods have evolved pleopodal lungs that provide access to the rich aerial supply of oxygen. However, isopods occupy conditions with wide and unpredictable thermal and oxygen gradients, suggesting that they might have evolved adaptive developmental plasticity in their respiratory organs to help meet metabolic demand over a wide range of oxygen conditions. To explore this plasticity, we conducted an experiment in which we reared common rough woodlice (Porcellio scaber) from eggs to maturation at different temperatures (15 and 22 °C) combined with different oxygen levels (10% and 22% O2). We sampled animals during development (only females) and then examined mature adults (both sexes). We compared woodlice between treatments with respect to the area of their pleopod exopodites (our proxy of lung size) and the shape of Bertalanffy's equations (our proxy of individual growth curves). Generally, males exhibited larger lungs than females relative to body size. Woodlice also grew relatively fast but achieved a decreased asymptotic body mass in response to warm conditions; the oxygen did not affect growth. Under hypoxia, growing females developed larger lungs compared to under normoxia, but only in the late stage of development. Among mature animals, this effect was present only in males. Woodlice reared under warm conditions had relatively small lungs, in both developing females (the effect was increased in relatively large females) and among mature males and females. Our results demonstrated that woodlice exhibit phenotypic plasticity in their lung size. We suggest that this plasticity helps woodlice equilibrate their gas exchange capacity to differences in the oxygen supply and metabolic demand along environmental temperature and oxygen gradients. The complex pattern of plasticity might indicate the effects of a balance between water conservation and oxygen uptake, which would be especially pronounced in mature females that need to generate an aqueous environment inside their brood pouch.

Corrigendum to "Decoding phenotypic screening: A comparative analysis of image representations" [Computat Struct Biotechnol J 23 (2024) 1181-1188].

[This corrects the article DOI: 10.1016/j.csbj.2024.02.022.].

Decoding phenotypic screening: A comparative analysis of image representations.

Biomedical imaging techniques such as high content screening (HCS) are valuable for drug discovery, but high costs limit their use to pharmaceutical companies. To address this issue, The JUMP-CP consortium released a massive open image dataset of chemical and genetic perturbations, providing a valuable resource for deep learning research. In this work, we aim to utilize the JUMP-CP dataset to develop a universal representation model for HCS data, mainly data generated using U2OS cells and CellPainting protocol, using supervised and self-supervised learning approaches. We propose an evaluation protocol that assesses their performance on mode of action and property prediction tasks using a popular phenotypic screening dataset. Results show that the self-supervised approach that uses data from multiple consortium partners provides representation that is more robust to batch effects whilst simultaneously achieving performance on par with standard approaches. Together with other conclusions, it provides recommendations on the training strategy of a representation model for HCS images.

Identifying Bacteria Species on Microscopic Polyculture Images Using Deep Learning.

Preliminary microbiological diagnosis usually relies on microscopic examination and, due to the routine culture and bacteriological examination, lasts up to 11 days. Hence, many deep learning methods based on microscopic images were recently introduced to replace the time-consuming bacteriological examination. They shorten the diagnosis by 1-2 days but still require iterative culture to obtain monoculture samples. In this work, we present a feasibility study for further shortening the diagnosis time by analyzing polyculture images. It is possible with multi-MIL, a novel multi-label classification method based on multiple instance learning. To evaluate our approach, we introduce a dataset containing microscopic images for all combinations of four considered bacteria species. We obtain ROC AUC above 0.9, proving the feasibility of the method and opening the path for future experiments with a larger number of species.

Deep learning models capture histological disease activity in Crohn's Disease and Ulcerative Colitis with high fidelity.

BACKGROUND AND AIMS: Histologic disease activity in Inflammatory Bowel Disease (IBD) is associated with clinical outcomes and is an important endpoint in drug development. We developed deep learning models for automating histological assessments in IBD. METHODS: Histology images of intestinal mucosa from phase 2 and phase 3 clinical trials in Crohn's disease (CD) and Ulcerative Colitis (UC) were used to train artificial intelligence (AI) models to predict the Global Histology Activity Score (GHAS) for CD and Geboes histopathology score for UC. Three AI methods were compared. AI models were evaluated on held-back testing sets and model predictions were compared against an expert central reader and five independent pathologists. RESULTS: The model based on multiple instance learning and the attention mechanism (SA-AbMILP) demonstrated the best performance among competing models. AI modeled GHAS and Geboes sub-grades matched central readings with moderate to substantial agreement, with accuracies ranging from 65% to 89%. Furthermore, the model was able to distinguish the presence and absence of pathology across four selected histological features with accuracies for colon, in both CD and UC, ranging from 87% to 94% and, for CD ileum, ranging from 76% to 83%. For both CD and UC, and across anatomical compartments (ileum and colon) in CD, comparable accuracies against central readings were found between the model assigned scores and scores by an independent set of pathologists. CONCLUSIONS: Deep learning models based upon GHAS and Geboes scoring systems were effective at distinguishing between the presence and absence of IBD microscopic disease activity.

Development of activated carbon for removal of pesticides from water: case study.

The work primarily concerns development of activated carbon dedicated for adsorption of pesticides from water prior directing it to the distribution system. We provide an information on research on important practical aspects related to research carried out to develop and to manufacture activated carbons. The paper concerns preliminary works on selection raw materials, a binder used for producing granulated adsorbent, activating gases, conditions of the production process, and others. The key attention in this research was paid to its target, i.e., industrial process to produce activated carbon revealing fulfilling required properties including satisfying adsorption of selected pesticides and meeting the requirements of companies dealing with a large-scale production of drinking water. Therefore, among others, the work includes considerations concerning such aspects like pore structure and specific surface area of the activated carbon, formation of granules that are the most demanded and thus preferred in an industrial practice form of activated carbons, and other aspects important from practical point of view. Using the results of our preliminary work, a batch of granular activated carbon was produced in industrial conditions. The obtained material was tested in terms of removing several pesticides at a water treatment plant operating on an industrial scale. During tests the concentration of acetochlor ESA was decreased from ca. 0.4 µg/l in raw water to below 0.1 µg/l. During 11 months of AC use specific surface area of adsorbent lowered significantly by 164 m2/g, and total pore volume declined from initial 0.56 cm3/g to 0.455 cm3/g. We discuss both a performance of the obtained activated carbon in a long-term removal of acetochlor and its derivatives from water and an effect of exploitation time on the removal efficiency. The explanations for the reduction in pesticide removal efficiency are also proposed and discussed.

Effect of the Granularity of Cubic Boron Nitride Vitrified Grinding Wheels on the Planar Technical Blades Sharpening Process.

The most widely used method for shaping technical blades is grinding with abrasive tools made of cubic boron nitride (cBN) grains and vitrified bond. The goal of this work was to determine the effect of grinding wheel grain size (cBN grain number according to FEPA standards: B126, B181 and B251), kinematics (grinding with the circumference, face and conical surface of the wheel) and feed rate (vf = 100; 150; 200 mm/min) on the effects of the grinding process evaluated by the cutting force of the blade after machining F, blade surface texture parameters (Sa, St, Smvr, Str, Sdq, Sdr and Sbi) as well as blade surface morphology. An analysis of output quantities showed that grinding wheels made of B181 cBN grains are most favorable for shaping planar technical blades of X39Cr13 steel in the grinding process.

Effect of Pro-Ecological Cooling and Lubrication Methods on the Sharpening Process of Planar Blades Used in Food Processing.

This work presents the results of an experimental study of the sharpening of planar technical blades used in the fish processing industry. Sharpening was carried out in the grinding process using several environmentally friendly methods of cooling and lubricating the machining zone (MQL method, CAG nozzle, hybrid method that is a combination of MQL and CAG methods, as well as WET flooding method as reference). The purpose of the research was to determine the possibility of reducing the negative environmental impact of the sharpening process of technical blades by minimizing the expenditure of coolant. The application of the MQL method and the hybrid MQL + CAG method provided a very good realization of the lubricating function so that the share of friction of dulled cutting vertices against the workpiece surface is reduced, which manifests itself in the reduction of the grinding force and the correlated grinding power. In the case of grinding under cooled compressed air delivery conditions, the average cutting force was as much as 91.6% higher (F = 22.63 N) compared to the result obtained for the most favorable flooding method, demonstrating the insufficient quality of the blade shaped under such conditions. A comprehensive comparison of test results on grinding power gain, cutting force and surface texture suggests that the most favorable sharpening results were obtained using the environmentally friendly MQL method of cooling and lubricating the grinding zone.

Comparison of Supervised and Self-Supervised Deep Representations Trained on Histological Images.

Self-supervised methods gain more and more attention, especially in the medical domain, where the number of labeled data is limited. They provide results on par or superior to their fully supervised competitors, yet the difference between information coded by both methods is unclear. This work introduces a novel comparison framework for explaining differences between supervised and self-supervised models using visual characteristics important to the human perceptual system. We apply this framework to models trained for Gleason score and conclude that self-supervised methods are more biased toward contrast and texture transformation than their supervised counterparts. At the same time, supervised methods code more information about the shape.

Methodology for Evaluating the Cutting Force of Planar Technical Blades Used in Flatfish Processing.

In the food industry, there are many varieties of technical blades with different contours as well as different cutting edge geometries. The evaluation of the ability of technical blades to separate (cut) animal tissues is not a simple task and is usually based on the evaluation of the cutting effects in a technological process. This paper presents a methodology for evaluating the cutting force of technical blades used in food processing. A specially made test stand with numerical control was used in the study. Its application enabled a comparison of cutting force values for four different cutting edge geometries of planar knives used in the skinning operation of flat fishes. A unique feature of the conducted research was the use of a relatively high cutting speed value of vf = 214 mm/s, which corresponded to the real conditions of this process carried out in the industry. Obtained test results allow unambiguously choosing the most advantageous variant of knife geometry from among four different variants used for the tests. The results showed a clear relationship between the cutting force value and the value of the tip angle of the blades tested: for blades with the lowest tip angle, the lowest cutting force values were obtained.

The Role of Observation-Measurement Methods in the Surface Characterization of X39Cr13 Stainless-Steel Cutting Blades Used in the Fish Processing Industry.

In the modern fish processing industry, flat fishes play an important role. They are processed into a final product in the form of a fillet during the skinning operation, which is carried out on machines operating in automated production lines. These machines are usually equipped with a single planar cutting blade or a few of such blades. The high-efficiency skinning and industrial conditions cause rapid wear of the cutting edge of the blade, which is detrimental to the quality of the final product. One of the forms of renewing the cutting ability of these types of tools is the regeneration carried out with the use of precise traverse surface grinding. The results of this process must be carefully verified for determining its correctness and possible optimization of its parameters. The main goal of this article was to characterize the usefulness of a number of observational and measuring methods to evaluate the results of the technical blade regeneration process. In this work, a number of contemporary observation-measurement methods such as optical microscopy (OM), scanning electron microscopy (SEM), energy-dispersive X-ray spectroscopy (EDS), optical profilometry (OP), and angle-resolved scattering (ARS), supported by image processing and analysis techniques, were analyzed. The authors focused on presenting the role of the abovementioned methods in the surface characterization of planar cutting blades made of X39Cr13 chromium martensitic stainless steel before and after the technological operation of flat-fish skinning. Additionally, the surface condition after the regeneration process carried out using the five-axis CNC (computerized numerical control) grinding machine was also assessed. Numerous results of surface observations, elemental composition microanalysis, high-accuracy surface microgeometry measurements, and quantitative and qualitative analysis confirming the possibility of using the proposed methods in the presented applications are presented.

Deep learning approach to describe and classify fungi microscopic images.

Preliminary diagnosis of fungal infections can rely on microscopic examination. However, in many cases, it does not allow unambiguous identification of the species due to their visual similarity. Therefore, it is usually necessary to use additional biochemical tests. That involves additional costs and extends the identification process up to 10 days. Such a delay in the implementation of targeted therapy may be grave in consequence as the mortality rate for immunosuppressed patients is high. In this paper, we apply a machine learning approach based on deep neural networks and bag-of-words to classify microscopic images of various fungi species. Our approach makes the last stage of biochemical identification redundant, shortening the identification process by 2-3 days, and reducing the cost of the diagnosis.

Label-free FTIR spectroscopy detects and visualizes the early stage of pulmonary micrometastasis seeded from breast carcinoma.

An application of FTIR spectroscopic imaging for the identification and visualization of early micrometastasis from breast cancer to lungs in a murine model is shown. Spectroscopic and histological examination is focused on lung cross-sections derived from animals at the early phase of metastasis (early micrometastasis, EM) as compared to healthy control (HC) and late phase of metastasis (advanced macrometastasis, AM) using murine model of metastatic breast cancer with 4T1 cells orthotopically inoculated. FTIR imaging allows for a detailed, objective and label-free differentiation and visualization of EM foci including large and small micrometastases as well as single cancer cells grouped in clusters. An effect of the EM phase on the entire lung tissue matrix as well as characteristic biochemical profiles for HC and advanced macrometastasis were determined from morphological and spectroscopic points of view. The extraordinary sensitivity of FTIR imaging toward EM detection and discrimination of AM borders confirms its applicability as a complementary tool for the histopathological assessment of the metastatic cancer progression.

Deep learning approach to bacterial colony classification.

In microbiology it is diagnostically useful to recognize various genera and species of bacteria. It can be achieved using computer-aided methods, which make the recognition processes more automatic and thus significantly reduce the time necessary for the classification. Moreover, in case of diagnostic uncertainty (the misleading similarity in shape or structure of bacterial cells), such methods can minimize the risk of incorrect recognition. In this article, we apply the state of the art method for texture analysis to classify genera and species of bacteria. This method uses deep Convolutional Neural Networks to obtain image descriptors, which are then encoded and classified with Support Vector Machine or Random Forest. To evaluate this approach and to make it comparable with other approaches, we provide a new dataset of images. DIBaS dataset (Digital Image of Bacterial Species) contains 660 images with 33 different genera and species of bacteria.