A Cell Segmentation/Tracking Tool Based on Machine Learning

A user-friendly machine learning approach for cardiac structures assessment

Background

Machine learning is increasingly being used to diagnose and treat various diseases, including cardiovascular diseases. Automatic image analysis can expedite tissue analysis and save time. However, using machine learning is limited among researchers due to the requirement of technical expertise. By offering extensible features through plugins and scripts, machine-learning platforms make these techniques more accessible to researchers with limited programming knowledge. The misuse of anabolic-androgenic steroids is prevalent, particularly among athletes and bodybuilders, and there is strong evidence of their detrimental effects on ventricular myocardial capillaries and muscle cells. However, most studies rely on qualitative data, which can lead to bias and limited reliability. We present a user-friendly approach using machine learning algorithms to measure the effects of exercise and anabolic-androgenic steroids on cardiac ventricular capillaries and myocytes in an experimental animal model.

Method

Male Wistar rats were divided into four groups (n = 28): control, exercise-only, anabolic-androgenic steroid-alone, and exercise with anabolic-androgenic steroid. Histopathological analysis of heart tissue was conducted, with images processed and analyzed using the Trainable Weka Segmentation plugin in Fiji software. Machine learning classifiers were trained to segment capillary and myocyte nuclei structures, enabling quantitative morphological measurements.

Results

Exercise significantly increased capillary density compared to other groups. However, in the exercise + anabolic-androgenic steroid group, steroid use counteracted this effect. Anabolic-androgenic steroid alone did not significantly impact capillary density compared to the control group. Additionally, the exercise group had a significantly shorter intercapillary distance than all other groups. Again, using steroids in the exercise + anabolic-androgenic steroid group diminished this positive effect.

Conclusion

Despite limited programming skills, researchers can use artificial intelligence techniques to investigate the adverse effects of anabolic steroids on the heart's vascular network and muscle cells. By employing accessible tools like machine learning algorithms and image processing software, histopathological images of capillary and myocyte structures in heart tissues can be analyzed.

Counter-on-chip for bacterial cell quantification, growth, and live-dead estimations

Quantifying bacterial cell numbers is crucial for experimental assessment and reproducibility, but the current technologies have limitations. The commonly used colony forming units (CFU) method causes a time delay in determining the actual numbers. Manual microscope counts are often error-prone for submicron bacteria. Automated systems are costly, require specialized knowledge, and are erroneous when counting smaller bacteria. In this study, we took a different approach by constructing three sequential generations (G1, G2, and G3) of counter-on-chip that accurately and timely count small particles and/or bacterial cells. We employed 2-photon polymerization (2PP) fabrication technology; and optimized the printing and molding process to produce high-quality, reproducible, accurate, and efficient counters. Our straightforward and refined methodology has shown itself to be highly effective in fabricating structures, allowing for the rapid construction of polydimethylsiloxane (PDMS)-based microfluidic devices. The G1 comprises three counting chambers with a depth of 20 µm, which showed accurate counting of 1 µm and 5 µm microbeads. G2 and G3 have eight counting chambers with depths of 20 µm and 5 µm, respectively, and can quickly and precisely count Escherichia coli cells. These systems are reusable, accurate, and easy to use (compared to CFU/ml). The G3 device can give (1) accurate bacterial counts, (2) serve as a growth chamber for bacteria, and (3) allow for live/dead bacterial cell estimates using staining kits or growth assay activities (live imaging, cell tracking, and counting). We made these devices out of necessity; we know no device on the market that encompasses all these features.

Escherichia coli cells are primed for survival before lethal antibiotic stress

Non-genetic factors can cause significant fluctuations in gene expression levels. Regardless of growing in a stable environment, this fluctuation leads to cell-to-cell variability in an isogenic population. This phenotypic heterogeneity allows a tiny subset of bacterial cells in a population called persister cells to tolerate long-term lethal antibiotic effects by entering into a non-dividing, metabolically repressed state. We occasionally noticed a high variation in persister levels, and to explore this, we tested clonal populations starting from a single cell using a modified Luria-Delbrück fluctuation test. Although we kept the conditions same, the diversity in persistence level among clones was relatively consistent: varying from ~60- to 100- and ~40- to 70-fold for ampicillin and apramycin, respectively. Then, we divided and diluted each clone to observe whether the same clone had comparable persister levels for more than one generation. Replicates had similar persister levels even when clones were divided, diluted by 1:20, and allowed to grow for approximately five generations. This result explicitly shows a cellular memory passed on for generations and eventually lost when cells are diluted to 1:100 and regrown (>seven generations). Our result demonstrates (1) the existence of a small population prepared for stress ("primed cells") resulting in higher persister numbers; (2) the primed memory state is reproducible and transient, passed down for generations but eventually lost; and (3) a heterogeneous persister population is a result of a transiently primed reversible cell state and not due to a pre-existing genetic mutation. IMPORTANCE Antibiotics have been highly effective in treating lethal infectious diseases for almost a century. However, the increasing threat of antibiotic resistance is again causing these diseases to become life-threatening. The longer a bacteria can survive antibiotics, the more likely it is to develop resistance. Complicating matters is that non-genetic factors can allow bacterial cells with identical DNA to gain transient resistance (also known as persistence). Here, we show that a small fraction of the bacterial population called primed cells can pass down non-genetic information ("memory") to their offspring, enabling them to survive lethal antibiotics for a long time. However, this memory is eventually lost. These results demonstrate how bacteria can leverage differences among genetically identical cells formed through non-genetic factors to form primed cells with a selective advantage to survive antibiotics.

A review of open-source image analysis tools for mammalian cell culture: algorithms, features and implementations

Cell culture is undeniably important for multiple scientific applications, including pharmaceuticals, transplants, and cosmetics. However, cell culture involves multiple manual steps, such as regularly analyzing cell images for their health and morphology. Computer scientists have developed algorithms to automate cell imaging analysis, but they are not widely adopted by biologists, especially those lacking an interactive platform. To address the issue, we compile and review existing open-source cell image processing tools that provide interactive interfaces for management and prediction tasks. We highlight the prediction tools that can detect, segment, and track different mammalian cell morphologies across various image modalities and present a comparison of algorithms and unique features of these tools, whether they work locally or in the cloud. This would guide non-experts to determine which is best suited for their purposes and, developers to acknowledge what is worth further expansion. In addition, we provide a general discussion on potential implementations of the tools for a more extensive scope, which guides the reader to not restrict them to prediction tasks only. Finally, we conclude the article by stating new considerations for the development of interactive cell imaging tools and suggesting new directions for future research.

Recent advances in experimental design and data analysis to characterize prokaryotic motility

Bacterial motility plays a key role in important cell processes such as chemotaxis and biofilm formation, but is challenging to quantify due to the small size of the individual microorganisms and the complex interplay of biological and physical factors that influence motility phenotypes. Swimming, the first type of motility described in bacteria, still remains largely unquantified. Light microscopy has enabled qualitative characterization of swimming patterns seen in different strains, such as run and tumble, run-reverse-flick, run and slow, stop and coil, and push and pull, which has allowed for elucidation of the underlying physics. However, quantifying these behaviors (e.g., identifying run distances and speeds, turn angles and behavior by surfaces or cell-cell interactions) remains a challenging task. A qualitative and quantitative understanding of bacterial motility is needed to bridge the gap between experimentation, omics analysis, and bacterial motility theory. In this review, we discuss the strengths and limitations of how phase contrast microscopy, fluorescence microscopy, and digital holographic microscopy have been used to quantify bacterial motility. Approaches to automated software analysis, including cell recognition, tracking, and track analysis, are also discussed with a view to providing a guide for experimenters to setting up the appropriate imaging and analysis system for their needs.

Machine Learning Quantification of Amyloid Deposits in Histological Images of Ligamentum Flavum

Wild-type transthyretin amyloidosis (ATTRwt) is an underdiagnosed and potentially fatal disease. Interestingly, ATTRwt deposits have been found to deposit in the ligamentum flavum (LF) of patients with lumbar spinal stenosis before the development of systemic and cardiac amyloidosis. In order to study this phenomenon and its possible relationship with LF thickening and systemic amyloidosis, a precise method of quantifying amyloid deposits in histological slides of LF is critical. However, such a method is currently unavailable. Here, we present a machine learning quantification method with Trainable Weka Segmentation (TWS) to assess amyloid deposition in histological slides of LF. Images of ligamentum flavum specimens stained with Congo red are obtained from spinal stenosis patients undergoing laminectomies and confirmed to be positive for ATTRwt. Amyloid deposits in these specimens are classified and quantified by TWS through training the algorithm via user-directed annotations on images of LF. TWS can also be automated through exposure to a set of training images with user-directed annotations, and then applied] to a set of new images without additional annotations. Additional methods of color thresholding and manual segmentation are also used on these images for comparison to TWS. We develop the use of TWS in images of LF and demonstrate its potential for automated quantification. TWS is strongly correlated with manual segmentation in the training set of images with user-directed annotations (R = 0.98; p = 0.0033) as well as in the application set of images where TWS was automated (R = 0.94; p = 0.016). Color thresholding was weakly correlated with manual segmentation in the training set of images (R = 0.78; p = 0.12) and in the application set of images (R = 0.65; p = 0.23). TWS machine learning closely correlates with the gold-standard comparator of manual segmentation and outperforms the color thresholding method. This novel machine learning method to quantify amyloid deposition in histological slides of ligamentum flavum is a precise, objective, accessible, high throughput, and powerful tool that will hopefully pave the way towards future research and clinical applications.

Learning deep features for dead and living breast cancer cell classification without staining

Automated cell classification in cancer biology is a challenging topic in computer vision and machine learning research. Breast cancer is the most common malignancy in women that usually involves phenotypically diverse populations of breast cancer cells and an heterogeneous stroma. In recent years, automated microscopy technologies are allowing the study of live cells over extended periods of time, simplifying the task of compiling large image databases. For instance, there have been several studies oriented towards building machine learning systems capable of automatically classifying images of different cell types (i.e. motor neurons, stem cells). In this work we were interested in classifying breast cancer cells as live or dead, based on a set of automatically retrieved morphological characteristics using image processing techniques. Our hypothesis is that live-dead classification can be performed without any staining and using only bright-field images as input. We tackled this problem using the JIMT-1 breast cancer cell line that grows as an adherent monolayer. First, a vast image set composed by JIMT-1 human breast cancer cells that had been exposed to a chemotherapeutic drug treatment (doxorubicin and paclitaxel) or vehicle control was compiled. Next, several classifiers were trained based on well-known convolutional neural networks (CNN) backbones to perform supervised classification using labels obtained from fluorescence microscopy images associated with each bright-field image. Model performances were evaluated and compared on a large number of bright-field images. The best model reached an AUC = 0.941 for classifying breast cancer cells without treatment. Furthermore, it reached AUC = 0.978 when classifying breast cancer cells under drug treatment. Our results highlight the potential of machine learning and computational image analysis to build new diagnosis tools that benefit the biomedical field by reducing cost, time, and stimulating work reproducibility. More importantly, we analyzed the way our classifiers clusterize bright-field images in the learned high-dimensional embedding and linked these groups to salient visual characteristics in live-dead cell biology observed by trained experts.

Proteolytic Queues at ClpXP Increase Antibiotic Tolerance

Antibiotic tolerance is a widespread phenomenon that renders antibiotic treatments less effective and facilitates antibiotic resistance. Here we explore the role of proteases in antibiotic tolerance, short-term population survival of antibiotics, using queueing theory (i.e., the study of waiting lines), computational models, and a synthetic biology approach. Proteases are key cellular components that degrade proteins and play an important role in a multidrug tolerant subpopulation of cells, called persisters. We found that queueing at the protease ClpXP increases antibiotic tolerance ∼80 and ∼60 fold in an E. coli population treated with ampicillin and ciprofloxacin, respectively. There does not appear to be an effect on antibiotic persistence, which we distinguish from tolerance based on population decay. These results demonstrate that proteolytic queueing is a practical method to probe proteolytic activity in bacterial tolerance and related genes, while limiting the unintended consequences frequently caused by gene knockout and overexpression.