Holographic deep learning for rapid optical screening of anthrax spores

Artificial intelligence in bacterial diagnostics and antimicrobial susceptibility testing: Current advances and future prospects

Recently, artificial intelligence (AI) has emerged as a transformative tool, enhancing the speed, accuracy, and scalability of bacterial diagnostics. This review explores the role of AI in revolutionizing bacterial detection and antimicrobial susceptibility testing (AST) by leveraging machine learning models, including Random Forest, Support Vector Machines (SVM), and deep learning architectures such as Convolutional Neural Networks (CNNs) and transformers. The integration of AI into these methods promises to address the current limitations of traditional techniques, offering a path toward more efficient, accessible, and reliable diagnostic solutions. In particular, AI-based approaches have demonstrated significant potential in resource-limited settings by enabling cost-effective and portable diagnostic solutions, reducing dependency on specialized infrastructure, and facilitating remote bacterial detection through smartphone-integrated platforms and telemedicine applications. This review highlights AI's transformative role in automating data analysis, minimizing human error, and delivering real-time diagnostic results, ultimately improving patient outcomes and optimizing healthcare efficiency. In addition, we not only examine the current advances in machine learning and deep learning but also review their applications in plate counting, mass spectrometry, morphology-based and motion-based microscopic detection, holographic microscopy, colorimetric and fluorescence detection, electrochemical sensors, Raman and Surface-Enhanced Raman Spectroscopy (SERS), and Atomic Force Microscopy (AFM) for bacterial diagnostics and AST. Finally, we discuss the future directions and potential advancements in AI-driven bacterial diagnostics.

Artificial Intelligence Methods in Infection Biology Research

Despite unprecedented achievements, the domain-specific application of artificial intelligence (AI) in the realm of infection biology was still in its infancy just a couple of years ago. This is largely attributable to the proneness of the infection biology community to shirk quantitative techniques. The so-called "sorting machine" paradigm was prevailing at that time, meaning that AI applications were primarily confined to the automation of tedious laboratory tasks. However, fueled by the severe acute respiratory syndrome coronavirus 2 pandemic, AI-driven applications in infection biology made giant leaps beyond mere automation. Instead, increasingly sophisticated tasks were successfully tackled, thereby ushering in the transition to the "Swiss army knife" paradigm. Incentivized by the urgent need to subdue a raging pandemic, AI achieved maturity in infection biology and became a versatile tool. In this chapter, the maturation of AI in the field of infection biology from the "sorting machine" paradigm to the "Swiss army knife" paradigm is outlined. Successful applications are illustrated for the three data modalities in the domain, that is, images, molecular data, and language data, with a particular emphasis on disentangling host-pathogen interactions. Along the way, fundamental terminology mentioned in the same breath as AI is elaborated on, and relationships between the subfields these terms represent are established. Notably, in order to dispel the fears of infection biologists toward quantitative methodologies and lower the initial hurdle, this chapter features a hands-on guide on software installation, virtual environment setup, data preparation, and utilization of pretrained models at its very end.

Can deep neural networks work with amplitude and phase input of defocused images?

Deep neural network (DNN) models, particularly convolutional neural networks (CNNs), have demonstrated remarkable performance in biomedical image classification due to their ability to automatically learn features from large datasets. One common challenge in the preparation of large, microscopic datasets for DNN tasks is sample defocusing, potentially impairing the model performance. To handle defocusing, computational imaging, or specifically quantitative phase imaging (QPI), performs digital refocusing by using both the phase and the amplitude of the complex optical field. This leads us to investigate whether feeding the complex field into DNN would potentially address the defocusing problem as in-focus information is implicitly encoded in the complex field. In this paper, we assess the feasibility of employing neural networks to directly process full amplitude and phase data from a defocus plane without digital refocusing. Our specific focus lies in understanding the tolerance for defocus in image classification neural networks when amplitude and phase are taken as inputs. To accomplish this, we used Fourier ptychography microscopy (FPM) to acquire in-focus phase and amplitude images of two distinct object types - normal red blood cells and echinocytes. We then digitally propagate the complex field to generate progressively defocused images of the samples to serve as training and testing datasets for image classification neural networks. While the digitally defocused images contain sufficient information to recover the original in-focus images, we observed that current standard implementations of deep learning models are unable to effectively utilize the defocused field to distinguish between the two cell types. We conclude that the physical-model-based digital refocusing capability of QPI remains indispensable for overcoming defocusing issues in current standard DNN models.

Quantitative phase imaging techniques for measuring scattering properties of cells and tissues: a review-part II

Significance

Quantitative phase imaging (QPI) is a non-invasive, label-free technique that provides intrinsic information about the sample under study. Such information includes the structure, function, and dynamics of the sample. QPI overcomes the limitations of conventional fluorescence microscopy in terms of phototoxicity to the sample and photobleaching of the fluorophore. As such, the application of QPI in estimating the three-dimensional (3D) structure and dynamics is well-suited for a range of samples from intracellular organelles to highly scattering multicellular samples while allowing for longer observation windows.

Aim

We aim to provide a comprehensive review of 3D QPI and related phase-based measurement techniques along with a discussion of methods for the estimation of sample dynamics.

Approach

We present information collected from 106 publications that cover the theoretical description of 3D light scattering and the implementation of related measurement techniques for the study of the structure and dynamics of the sample. We conclude with a discussion of the applications of the reviewed techniques in the biomedical field.

Results

QPI has been successfully applied to 3D sample imaging. The scattering-based contrast provides measurements of intrinsic quantities of the sample that are indicative of disease state, stage of growth, or overall dynamics.

Conclusions

We reviewed state-of-the-art QPI techniques for 3D imaging and dynamics estimation of biological samples. Both theoretical and experimental aspects of various techniques were discussed. We also presented the applications of the discussed techniques as applied to biomedicine and biology research.

Quantitative phase imaging techniques for measuring scattering properties of cells and tissues: a review-part I

Significance

Quantitative phase imaging (QPI) techniques offer intrinsic information about the sample of interest in a label-free, noninvasive manner and have an enormous potential for wide biomedical applications with negligible perturbations to the natural state of the sample in vitro.

Aim

We aim to present an in-depth review of the scattering formulation of light-matter interactions as applied to biological samples such as cells and tissues, discuss the relevant quantitative phase measurement techniques, and present a summary of various reported applications.

Approach

We start with scattering theory and scattering properties of biological samples followed by an exploration of various microscopy configurations for 2D QPI for measurement of structure and dynamics.

Results

We reviewed 157 publications and presented a range of QPI techniques and discussed suitable applications for each. We also presented the theoretical frameworks for phase reconstruction associated with the discussed techniques and highlighted their domains of validity.

Conclusions

We provide detailed theoretical as well as system-level information for a wide range of QPI techniques. Our study can serve as a guideline for new researchers looking for an exhaustive literature review of QPI methods and relevant applications.

Holotomography and atomic force microscopy: a powerful combination to enhance cancer, microbiology and nanotoxicology research

Modern imaging strategies are paramount to studying living systems such as cells, bacteria, and fungi and their response to pathogens, toxicants, and nanomaterials (NMs) as modulated by exposure and environmental factors. The need to understand the processes and mechanisms of damage, healing, and cell survivability of living systems continues to motivate the development of alternative imaging strategies. Of particular interest is the use of label-free techniques (microscopy procedures that do not require sample staining) that minimize interference of biological processes by foreign marking substances and reduce intense light exposure and potential photo-toxicity effects. This review focuses on the synergic capabilities of atomic force microscopy (AFM) as a well-developed and robust imaging strategy with demonstrated applications to unravel intimate details in biomedical applications, with the label-free, fast, and enduring Holotomographic Microscopy (HTM) strategy. HTM is a technique that combines holography and tomography using a low intensity continuous illumination laser to investigate (quantitatively and non-invasively) cells, microorganisms, and thin tissue by generating three-dimensional (3D) images and monitoring in real-time inner morphological changes. We first review the operating principles that form the basis for the complementary details provided by these techniques regarding the surface and internal information provided by HTM and AFM, which are essential and complimentary for the development of several biomedical areas studying the interaction mechanisms of NMs with living organisms. First, AFM can provide superb resolution on surface morphology and biomechanical characterization. Second, the quantitative phase capabilities of HTM enable superb modeling and quantification of the volume, surface area, protein content, and mass density of the main components of cells and microorganisms, including the morphology of cells in microbiological systems. These capabilities result from directly quantifying refractive index changes without requiring fluorescent markers or chemicals. As such, HTM is ideal for long-term monitoring of living organisms in conditions close to their natural settings. We present a case-based review of the principal uses of both techniques and their essential contributions to nanomedicine and nanotoxicology (study of the harmful effects of NMs in living organisms), emphasizing cancer and infectious disease control. The synergic impact of the sequential use of these complementary strategies provides a clear drive for adopting these techniques as interdependent fundamental tools.

EVATOM: an optical, label-free, machine learning assisted embryo health assessment tool

The combination of a good quality embryo and proper maternal health factors promise higher chances of a successful in vitro fertilization (IVF) procedure leading to clinical pregnancy and live birth. Of these two factors, selection of a good embryo is a controllable aspect. The current gold standard in clinical practice is visual assessment of an embryo based on its morphological appearance by trained embryologists. More recently, machine learning has been incorporated into embryo selection "packages". Here, we report EVATOM: a machine-learning assisted embryo health assessment tool utilizing an optical quantitative phase imaging technique called artificial confocal microscopy (ACM). We present a label-free nucleus detection method with, to the best of our knowledge, novel quantitative embryo health biomarkers. Two viability assessment models are presented for grading embryos into two classes: healthy/intermediate (H/I) or sick (S) class. The models achieve a weighted F1 score of 1.0 and 0.99 respectively on the in-distribution test set of 72 fixed embryos and a weighted F1 score of 0.9 and 0.95 respectively on the out-of-distribution test dataset of 19 time-instances from 8 live embryos.

Towards ultrafast quantitative phase imaging via differentiable microscopy [Invited]

With applications ranging from metabolomics to histopathology, quantitative phase microscopy (QPM) is a powerful label-free imaging modality. Despite significant advances in fast multiplexed imaging sensors and deep-learning-based inverse solvers, the throughput of QPM is currently limited by the pixel-rate of the image sensors. Complementarily, to improve throughput further, here we propose to acquire images in a compressed form so that more information can be transferred beyond the existing hardware bottleneck of the image sensor. To this end, we present a numerical simulation of a learnable optical compression-decompression framework that learns content-specific features. The proposed differentiable quantitative phase microscopy (∂-QPM) first uses learnable optical processors as image compressors. The intensity representations produced by these optical processors are then captured by the imaging sensor. Finally, a reconstruction network running on a computer decompresses the QPM images post aquisition. In numerical experiments, the proposed system achieves compression of × 64 while maintaining the SSIM of ∼0.90 and PSNR of ∼30 dB on cells. The results demonstrated by our experiments open up a new pathway to QPM systems that may provide unprecedented throughput improvements.

Toward the novel AI tasks in infection biology

Machine learning and artificial intelligence (AI) are becoming more common in infection biology laboratories around the world. Yet, as they gain traction in research, novel frontiers arise. Novel artificial intelligence algorithms are capable of addressing advanced tasks like image generation and question answering. However, similar algorithms can prove useful in addressing advanced questions in infection biology like prediction of host-pathogen interactions or inferring virus protein conformations. Addressing such tasks requires large annotated data sets, which are often scarce in biomedical research. In this review, I bring together several successful examples where such tasks were addressed. I underline the importance of formulating novel AI tasks in infection biology accompanied by freely available benchmark data sets to address these tasks. Furthermore, I discuss the current state of the field and potential future trends. I argue that one such trend involves AI tools becoming more versatile.

Real-time simultaneous refractive index and thickness mapping of sub-cellular biology at the diffraction limit

Mapping the cellular refractive index (RI) is a central task for research involving the composition of microorganisms and the development of models providing automated medical screenings with accuracy beyond 95%. These models require significantly enhancing the state-of-the-art RI mapping capabilities to provide large amounts of accurate RI data at high throughput. Here, we present a machine-learning-based technique that obtains a biological specimen's real-time RI and thickness maps from a single image acquired with a conventional color camera. This technology leverages a suitably engineered nanostructured membrane that stretches a biological analyte over its surface and absorbs transmitted light, generating complex reflection spectra from each sample point. The technique does not need pre-existing sample knowledge. It achieves 10-4 RI sensitivity and sub-nanometer thickness resolution on diffraction-limited spatial areas. We illustrate practical application by performing sub-cellular segmentation of HCT-116 colorectal cancer cells, obtaining complete three-dimensional reconstruction of the cellular regions with a characteristic length of 30 μm. These results can facilitate the development of real-time label-free technologies for biomedical studies on microscopic multicellular dynamics.

On the use of deep learning for phase recovery

Phase recovery (PR) refers to calculating the phase of the light field from its intensity measurements. As exemplified from quantitative phase imaging and coherent diffraction imaging to adaptive optics, PR is essential for reconstructing the refractive index distribution or topography of an object and correcting the aberration of an imaging system. In recent years, deep learning (DL), often implemented through deep neural networks, has provided unprecedented support for computational imaging, leading to more efficient solutions for various PR problems. In this review, we first briefly introduce conventional methods for PR. Then, we review how DL provides support for PR from the following three stages, namely, pre-processing, in-processing, and post-processing. We also review how DL is used in phase image processing. Finally, we summarize the work in DL for PR and provide an outlook on how to better use DL to improve the reliability and efficiency of PR. Furthermore, we present a live-updating resource ( https://github.com/kqwang/phase-recovery ) for readers to learn more about PR.

Artificial intelligence-enabled quantitative phase imaging methods for life sciences

Quantitative phase imaging, integrated with artificial intelligence, allows for the rapid and label-free investigation of the physiology and pathology of biological systems. This review presents the principles of various two-dimensional and three-dimensional label-free phase imaging techniques that exploit refractive index as an intrinsic optical imaging contrast. In particular, we discuss artificial intelligence-based analysis methodologies for biomedical studies including image enhancement, segmentation of cellular or subcellular structures, classification of types of biological samples and image translation to furnish subcellular and histochemical information from label-free phase images. We also discuss the advantages and challenges of artificial intelligence-enabled quantitative phase imaging analyses, summarize recent notable applications in the life sciences, and cover the potential of this field for basic and industrial research in the life sciences.

Quantitative phase imaging for characterization of single cell growth dynamics

Quantitative phase imaging (QPI) has emerged as an indispensable tool in the field of biomedicine, offering the ability to obtain quantitative maps of phase changes due to optical path length delays without the need for contrast agents. These maps provide valuable information about cellular morphology and dynamics, unperturbed by the introduction of exogenous substances. In this review, a summary of recent studies that have focused on elucidating the growth dynamics of individual cells using QPI is presented. Specifically, investigations into cellular changes occurring during mitosis, the differentiation of cellular organelles, the assessment of distinct cell death processes (i.e., apoptosis, necrosis, and oncosis) and the precise measurement of live cell temperature are explored. Furthermore, the captivating applications of QPI in theragnostics, where its potential for transformative impact is prominently showcased, are highlighted. Finally, the challenges that need to be overcome for its wider adoption and successful integration into biomedical research are outlined.

Machine learning assisted health viability assay for mouse embryos with artificial confocal microscopy (ACM)

The combination of a good quality embryo and proper maternal health factors promise higher chances of a successful in vitro fertilization (IVF) procedure leading to clinical pregnancy and live birth. Of these two factors, selection of a good embryo is a controllable aspect. The current gold standard in clinical practice is visual assessment of an embryo based on its morphological appearance by trained embryologists. More recently, machine learning has been incorporated into embryo selection "packages". Here, we report a machine-learning assisted embryo health assessment tool utilizing a quantitative phase imaging technique called artificial confocal microscopy (ACM). We present a label-free nucleus detection method with novel quantitative embryo health biomarkers. Two viability assessment models are presented for grading embryos into two classes: healthy/intermediate (H/I) or sick (S) class. The models achieve a weighted F1 score of 1.0 and 0.99 respectively on the in-distribution test set of 72 fixed embryos and a weighted F1 score of 0.9 and 0.95 respectively on the out-of-distribution test dataset of 19 time-instances from 8 live embryos.

Viewing life without labels under optical microscopes

Optical microscopes today have pushed the limits of speed, quality, and observable space in biological specimens revolutionizing how we view life today. Further, specific labeling of samples for imaging has provided insight into how life functions. This enabled label-based microscopy to percolate and integrate into mainstream life science research. However, the use of labelfree microscopy has been mostly limited, resulting in testing for bio-application but not bio-integration. To enable bio-integration, such microscopes need to be evaluated for their timeliness to answer biological questions uniquely and establish a long-term growth prospect. The article presents key label-free optical microscopes and discusses their integrative potential in life science research for the unperturbed analysis of biological samples.

Multi-Class Classification and Multi-Output Regression of Three-Dimensional Objects Using Artificial Intelligence Applied to Digital Holographic Information

Digital holographically sensed 3D data processing, which is useful for AI-based vision, is demonstrated. Three prominent methods of learning from datasets such as sensed holograms, computationally retrieved intensity and phase from holograms forming concatenated intensity-phase (whole information) images, and phase-only images (depth information) were utilized for the proposed multi-class classification and multi-output regression tasks of the chosen 3D objects in supervised learning. Each dataset comprised 2268 images obtained from the chosen eighteen 3D objects. The efficacy of our approaches was validated on experimentally generated digital holographic data then further quantified and compared using specific evaluation matrices. The machine learning classifiers had better AUC values for different classes on the holograms and whole information datasets compared to the CNN, whereas the CNN had a better performance on the phase-only image dataset compared to these classifiers. The MLP regressor was found to have a stable prediction in the test and validation sets with a fixed EV regression score of 0.00 compared to the CNN, the other regressors for holograms, and the phase-only image datasets, whereas the RF regressor showed a better performance in the validation set for the whole information dataset with a fixed EV regression score of 0.01 compared to the CNN and other regressors.

Highly sensitive quantitative phase microscopy and deep learning aided with whole genome sequencing for rapid detection of infection and antimicrobial resistance

Current state-of-the-art infection and antimicrobial resistance (AMR) diagnostics are based on culture-based methods with a detection time of 48-96 h. Therefore, it is essential to develop novel methods that can do real-time diagnoses. Here, we demonstrate that the complimentary use of label-free optical assay with whole-genome sequencing (WGS) can enable rapid diagnosis of infection and AMR. Our assay is based on microscopy methods exploiting label-free, highly sensitive quantitative phase microscopy (QPM) followed by deep convolutional neural networks-based classification. The workflow was benchmarked on 21 clinical isolates from four WHO priority pathogens that were antibiotic susceptibility tested, and their AMR profile was determined by WGS. The proposed optical assay was in good agreement with the WGS characterization. Accurate classification based on the gram staining (100% recall for gram-negative and 83.4% for gram-positive), species (98.6%), and resistant/susceptible type (96.4%), as well as at the individual strain level (100% sensitivity in predicting 19 out of the 21 strains, with an overall accuracy of 95.45%). The results from this initial proof-of-concept study demonstrate the potential of the QPM assay as a rapid and first-stage tool for species, strain-level classification, and the presence or absence of AMR, which WGS can follow up for confirmation. Overall, a combined workflow with QPM and WGS complemented with deep learning data analyses could, in the future, be transformative for detecting and identifying pathogens and characterization of the AMR profile and antibiotic susceptibility.

Fourier ptychographic microscopy with untrained deep neural network priors

We propose a physics-assisted deep neural network scheme in Fourier ptychographic microscopy (FPM) using untrained deep neural network priors (FPMUP) to achieve a high-resolution image reconstruction from multiple low-resolution images. Unlike the traditional training type of deep neural network that requires a large labelled dataset, this proposed scheme does not require training and instead outputs the high-resolution image by optimizing the parameters of neural networks to fit the experimentally measured low-resolution images. Besides the amplitude and phase of the sample function, another two parallel neural networks that generate the general pupil function and illumination intensity factors are incorporated into the carefully designed neural networks, which effectively improves the image quality and robustness when both the aberration and illumination intensity fluctuation are present in FPM. Reconstructions using simulated and experimental datasets are demonstrated, showing that the FPMUP scheme has better image quality than the traditional iterative algorithms, especially for the phase recovery, but at the expense of increasing computational cost. Most importantly, it is found that the FPMUP scheme can predict the Fourier spectrum of the sample outside synthetic aperture of FPM and thus eliminate the ringing effect of the recovered images due to the spectral truncation. Inspired by deep image prior in the field of image processing, we may impute the expansion of Fourier spectrums to the deep prior rooted in the architecture of the careful designed four parallel deep neural networks. We envisage that the resolution of FPM will be further enhanced if the Fourier spectrum of the sample outside the synthetic aperture of FPM is accurately predicted.

COVID-19 screening with digital holographic microscopy using intra-patient probability functions of spatio-temporal bio-optical attributes

We present an automated method for COVID-19 screening using the intra-patient population distributions of bio-optical attributes extracted from digital holographic microscopy reconstructed red blood cells. Whereas previous approaches have aimed to identify infection by classifying individual cells, here, we propose an approach to incorporate the attribute distribution information from the population of a given human subjects' cells into our classification scheme and directly classify subjects at the patient level. To capture the intra-patient distribution information in a generalized way, we propose an approach based on the Bag-of-Features (BoF) methodology to transform histograms of bio-optical attribute distributions into feature vectors for classification via a linear support vector machine. We compare our approach with simpler classifiers directly using summary statistics such as mean, standard deviation, skewness, and kurtosis of the distributions. We also compare to a k-nearest neighbor classifier using the Kolmogorov-Smirnov distance as a distance metric between the attribute distributions of each subject. We lastly compare our approach to previously published methods for classification of individual red blood cells. In each case, the methodology proposed in this paper provides the highest patient classification performance, correctly classifying 22 out of 24 individuals and achieving 91.67% classification accuracy with 90.00% sensitivity and 92.86% specificity. The incorporation of distribution information for classification additionally led to the identification of a singular temporal-based bio-optical attribute capable of highly accurate patient classification. To the best of our knowledge, this is the first report of a machine learning approach using the intra-patient probability distribution information of bio-optical attributes obtained from digital holographic microscopy for disease screening.

Automated sickle cell disease identification in human red blood cells using a lensless single random phase encoding biosensor and convolutional neural networks

We present a compact, field portable, lensless, single random phase encoding biosensor for automated classification between healthy and sickle cell disease human red blood cells. Microscope slides containing 3 µl wet mounts of whole blood samples from healthy and sickle cell disease afflicted human donors are input into a lensless single random phase encoding (SRPE) system for disease identification. A partially coherent laser source (laser diode) illuminates the cells under inspection wherein the object complex amplitude propagates to and is pseudorandomly encoded by a diffuser, then the intensity of the diffracted complex waveform is captured by a CMOS image sensor. The recorded opto-biological signatures are transformed using local binary pattern map generation during preprocessing then input into a pretrained convolutional neural network for classification between healthy and disease-states. We further provide analysis that compares the performance of several neural network architectures to optimize our classification strategy. Additionally, we assess the performance and computational savings of classifying on subsets of the opto-biological signatures with substantially reduced dimensionality, including one dimensional cropping of the recorded signatures. To the best of our knowledge, this is the first report of a lensless SRPE biosensor for human disease identification. As such, the presented approach and results can be significant for low-cost disease identification both in the field and for healthcare systems in developing countries which suffer from constrained resources.

Quantitative Phase Imaging: Recent Advances and Expanding Potential in Biomedicine

Quantitative phase imaging (QPI) is a label-free, wide-field microscopy approach with significant opportunities for biomedical applications. QPI uses the natural phase shift of light as it passes through a transparent object, such as a mammalian cell, to quantify biomass distribution and spatial and temporal changes in biomass. Reported in cell studies more than 60 years ago, ongoing advances in QPI hardware and software are leading to numerous applications in biology, with a dramatic expansion in utility over the past two decades. Today, investigations of cell size, morphology, behavior, cellular viscoelasticity, drug efficacy, biomass accumulation and turnover, and transport mechanics are supporting studies of development, physiology, neural activity, cancer, and additional physiological processes and diseases. Here, we review the field of QPI in biology starting with underlying principles, followed by a discussion of technical approaches currently available or being developed, and end with an examination of the breadth of applications in use or under development. We comment on strengths and shortcomings for the deployment of QPI in key biomedical contexts and conclude with emerging challenges and opportunities based on combining QPI with other methodologies that expand the scope and utility of QPI even further.

A Deep-Learning Based System for Rapid Genus Identification of Pathogens under Hyperspectral Microscopic Images

Infectious diseases have always been a major threat to the survival of humanity. Additionally, they bring an enormous economic burden to society. The conventional methods for bacteria identification are expensive, time-consuming and laborious. Therefore, it is of great importance to automatically rapidly identify pathogenic bacteria in a short time. Here, we constructed an AI-assisted system for automating rapid bacteria genus identification, combining the hyperspectral microscopic technology and a deep-learning-based algorithm Buffer Net. After being trained and validated in the self-built dataset, which consists of 11 genera with over 130,000 hyperspectral images, the accuracy of the algorithm could achieve 94.9%, which outperformed 1D-CNN, 2D-CNN and 3D-ResNet. The AI-assisted system we developed has great potential in assisting clinicians in identifying pathogenic bacteria at the single-cell level with high accuracy in a cheap, rapid and automatic way. Since the AI-assisted system can identify the pathogenic genus rapidly (about 30 s per hyperspectral microscopic image) at the single-cell level, it can shorten the time or even eliminate the demand for cultivating. Additionally, the system is user-friendly for novices.

Rapid species identification of pathogenic bacteria from a minute quantity exploiting three-dimensional quantitative phase imaging and artificial neural network

The healthcare industry is in dire need of rapid microbial identification techniques for treating microbial infections. Microbial infections are a major healthcare issue worldwide, as these widespread diseases often develop into deadly symptoms. While studies have shown that an early appropriate antibiotic treatment significantly reduces the mortality of an infection, this effective treatment is difficult to practice. The main obstacle to early appropriate antibiotic treatments is the long turnaround time of the routine microbial identification, which includes time-consuming sample growth. Here, we propose a microscopy-based framework that identifies the pathogen from single to few cells. Our framework obtains and exploits the morphology of the limited sample by incorporating three-dimensional quantitative phase imaging and an artificial neural network. We demonstrate the identification of 19 bacterial species that cause bloodstream infections, achieving an accuracy of 82.5% from an individual bacterial cell or cluster. This performance, comparable to that of the gold standard mass spectroscopy under a sufficient amount of sample, underpins the effectiveness of our framework in clinical applications. Furthermore, our accuracy increases with multiple measurements, reaching 99.9% with seven different measurements of cells or clusters. We believe that our framework can serve as a beneficial advisory tool for clinicians during the initial treatment of infections.

Resolution and Contrast Enhancement for Lensless Digital Holographic Microscopy and Its Application in Biomedicine

An important imaging technique in biomedicine, the conventional optical microscopy relies on relatively complicated and bulky lens and alignment mechanics. Based on the Gabor holography, the lensless digital holographic microscopy has the advantages of light weight and low cost. It has developed rapidly and received attention in many fields. However, the finite pixel size at the sensor plane limits the spatial resolution. In this study, we first review the principle of lensless digital holography, then go over some methods to improve image contrast and discuss the methods to enhance the image resolution of the lensless holographic image. Moreover, the applications of lensless digital holographic microscopy in biomedicine are reviewed. Finally, we look forward to the future development and prospect of lensless digital holographic technology.

Understanding the Research Landscape of Deep Learning in Biomedical Science: Scientometric Analysis

BACKGROUND

Advances in biomedical research using deep learning techniques have generated a large volume of related literature. However, there is a lack of scientometric studies that provide a bird's-eye view of them. This absence has led to a partial and fragmented understanding of the field and its progress.

OBJECTIVE

This study aimed to gain a quantitative and qualitative understanding of the scientific domain by analyzing diverse bibliographic entities that represent the research landscape from multiple perspectives and levels of granularity.

METHODS

We searched and retrieved 978 deep learning studies in biomedicine from the PubMed database. A scientometric analysis was performed by analyzing the metadata, content of influential works, and cited references.

RESULTS

In the process, we identified the current leading fields, major research topics and techniques, knowledge diffusion, and research collaboration. There was a predominant focus on applying deep learning, especially convolutional neural networks, to radiology and medical imaging, whereas a few studies focused on protein or genome analysis. Radiology and medical imaging also appeared to be the most significant knowledge sources and an important field in knowledge diffusion, followed by computer science and electrical engineering. A coauthorship analysis revealed various collaborations among engineering-oriented and biomedicine-oriented clusters of disciplines.

CONCLUSIONS

This study investigated the landscape of deep learning research in biomedicine and confirmed its interdisciplinary nature. Although it has been successful, we believe that there is a need for diverse applications in certain areas to further boost the contributions of deep learning in addressing biomedical research problems. We expect the results of this study to help researchers and communities better align their present and future work.

Deep learning-based image processing in optical microscopy

Optical microscopy has emerged as a key driver of fundamental research since it provides the ability to probe into imperceptible structures in the biomedical world. For the detailed investigation of samples, a high-resolution image with enhanced contrast and minimal damage is preferred. To achieve this, an automated image analysis method is preferable over manual analysis in terms of both speed of acquisition and reduced error accumulation. In this regard, deep learning (DL)-based image processing can be highly beneficial. The review summarises and critiques the use of DL in image processing for the data collected using various optical microscopic techniques. In tandem with optical microscopy, DL has already found applications in various problems related to image classification and segmentation. It has also performed well in enhancing image resolution in smartphone-based microscopy, which in turn enablse crucial medical assistance in remote places.

Graphical abstract

A review of microscopic cell imaging and neural network recognition for synergistic cyanobacteria identification and enumeration

Real-time cyanobacteria/algal monitoring is a valuable tool for early detection of harmful algal blooms, water treatment efficacy evaluation, and assists tailored water quality risk assessments by considering taxonomy and cell counts. This review evaluates and proposes a synergistic approach using neural network image recognition and microscopic imaging devices by first evaluating published literature for both imaging microscopes and image recognition. Quantitative phase imaging was considered the most promising of the investigated imaging techniques due to the provision of enhanced information relative to alternatives. This information provides significant value to image recognition neural networks, such as the convolutional neural networks discussed within this review. Considering published literature, a cyanobacteria monitoring system and corresponding image processing workflow using in situ sample collection buoys and on-shore sample processing was proposed. This system can be implemented using commercially available equipment to facilitate accurate, real-time water quality monitoring.

COVID-19 detection from red blood cells using highly comparative time-series analysis (HCTSA) in digital holographic microscopy

We present an automated method for COVID-19 screening based on reconstructed phase profiles of red blood cells (RBCs) and a highly comparative time-series analysis (HCTSA). Video digital holographic data -was obtained using a compact, field-portable shearing microscope to capture the temporal fluctuations and spatio-temporal dynamics of live RBCs. After numerical reconstruction of the digital holographic data, the optical volume is calculated at each timeframe of the reconstructed data to produce a time-series signal for each cell in our dataset. Over 6000 features are extracted on the time-varying optical volume sequences using the HCTSA to quantify the spatio-temporal behavior of the RBCs, then a linear support vector machine is used for classification of individual RBCs. Human subjects are then classified for COVID-19 based on the consensus of their cells' classifications. The proposed method is tested on a dataset of 1472 RBCs from 24 human subjects (10 COVID-19 positive, 14 healthy) collected at UConn Health Center. Following a cross-validation procedure, our system achieves 82.13% accuracy, with 92.72% sensitivity, and 73.21% specificity (area under the receiver operating characteristic curve: 0.8357). Furthermore, the proposed system resulted in 21 out of 24 human subjects correctly labeled. To the best of our knowledge this is the first report of a highly comparative time-series analysis using digital holographic microscopy data.

Detection of Cytopathic Effects Induced by Influenza, Parainfluenza, and Enterovirus Using Deep Convolution Neural Network

The isolation of a virus using cell culture to observe its cytopathic effects (CPEs) is the main method for identifying the viruses in clinical specimens. However, the observation of CPEs requires experienced inspectors and excessive time to inspect the cell morphology changes. In this study, we utilized artificial intelligence (AI) to improve the efficiency of virus identification. After some comparisons, we used ResNet-50 as a backbone with single and multi-task learning models to perform deep learning on the CPEs induced by influenza, enterovirus, and parainfluenza. The accuracies of the single and multi-task learning models were 97.78% and 98.25%, respectively. In addition, the multi-task learning model increased the accuracy of the single model from 95.79% to 97.13% when only a few data of the CPEs induced by parainfluenza were provided. We modified both models by inserting a multiplexer and de-multiplexer layer, respectively, to increase the correct rates for known cell lines. In conclusion, we provide a deep learning structure with ResNet-50 and the multi-task learning model and show an excellent performance in identifying virus-induced CPEs.

A deep-learning-based workflow to deal with the defocusing problem in high-throughput experiments

The increasing throughput of experiments in biomaterials research makes automatic techniques more and more necessary. Among all the characterization methods, microscopy makes fundamental contributions to biomaterials science where precisely focused images are the basis of related research. Although automatic focusing has been widely applied in all kinds of microscopes, defocused images can still be acquired now and then due to factors including background noises of materials and mechanical errors. Herein, we present a deep-learning-based method for the automatic sorting and reconstruction of defocused cell images. First, the defocusing problem is illustrated on a high-throughput cell microarray. Then, a comprehensive dataset of phase-contrast images captured from varied conditions containing multiple cell types, magnifications, and substrate materials is prepared to establish and test our method. We obtain high accuracy of over 0.993 on the dataset using a simple network architecture that requires less than half of the training time compared with the classical ResNetV2 architecture. Moreover, the subcellular-level reconstruction of heavily defocused cell images is achieved with another architecture. The applicability of the established workflow in practice is finally demonstrated on the high-throughput cell microarray. The intelligent workflow does not require a priori knowledge of focusing algorithms, possessing widespread application value in cell experiments concerning high-throughput or time-lapse imaging.

•

An automatic workflow for the sorting and reconstruction of defocused images.

•

Representative data containing varied cells, substrate materials, and magnifications.

•

Effectiveness of the workflow in high-throughput experiments.

•

Publicly accessible image datasets and codes.

Deep learning for digital holography: a review

Recent years have witnessed the unprecedented progress of deep learning applications in digital holography (DH). Nevertheless, there remain huge potentials in how deep learning can further improve performance and enable new functionalities for DH. Here, we survey recent developments in various DH applications powered by deep learning algorithms. This article starts with a brief introduction to digital holographic imaging, then summarizes the most relevant deep learning techniques for DH, with discussions on their benefits and challenges. We then present case studies covering a wide range of problems and applications in order to highlight research achievements to date. We provide an outlook of several promising directions to widen the use of deep learning in various DH applications.

Image learning to accurately identify complex mixture components

The study of complex mixtures is very important for exploring the evolution of natural phenomena, but the complexity of the mixtures greatly increases the difficulty of material information extraction. Image perception-based machine-learning techniques have the ability to cope with this problem in a data-driven way. Herein, we report a 2D-spectral imaging method to collect matter information from mixture components, and the obtained feature images can be easily provided to deep convolutional neural networks (CNNs) for establishing a spectral network. The results demonstrated that a single CNN trained end-to-end from the proposed images can directly accomplish synchronous measurement of multi-component samples using only raw pixels as inputs. Our strategy has some innate advantages, such as fast data acquisition, low cost, and simple chemical treatment, suggesting that it can be extensively applied in many fields, including environmental science, biology, medicine, and chemistry.

Volumetric monitoring of airborne particulate matter concentration using smartphone-based digital holographic microscopy and deep learning

Airborne particulate matter (PM) has become a global environmental issue. This PM has harmful effects on public health and precision industries. Conventional air-quality monitoring methods usually utilize expensive equipment, and they are cumbersome to handle for accurate and high throughput measurements. In addition, commercial particle counters have technical limitations in high-concentration measurement, and data fluctuations are induced during air sampling. In this study, a novel smartphone-based technique for monitoring airborne PM concentrations was developed using smartphone-based digital holographic microscopy (S-DHM) and deep learning network called Holo-SpeckleNet. Holographic speckle images of various PM concentrations were recorded by the S-DHM system. The recorded speckle images and the corresponding ground truth PM concentrations were used to train deep learning algorithms consisting of a deep autoencoder and regression layers. The performance of the proposed smartphone-based PM monitoring technique was validated through hyperparameter optimization. The developed S-DHM integrated with Holo-SpeckleNet can be smartly and effectively utilized for portable PM monitoring and safety alarm provision under perilous environmental conditions.

Digital holographic deep learning of red blood cells for field-portable, rapid COVID-19 screening

Rapid screening of red blood cells for active infection of COVID-19 is presented using a compact and field-portable, 3D-printed shearing digital holographic microscope. Video holograms of thin blood smears are recorded, individual red blood cells are segmented for feature extraction, then a bi-directional long short-term memory network is used to classify between healthy and COVID positive red blood cells based on their spatiotemporal behavior. Individuals are then classified based on the simple majority of their cells' classifications. The proposed system may be beneficial for under-resourced healthcare systems. To the best of our knowledge, this is the first report of digital holographic microscopy for rapid screening of COVID-19.

Point-of-care diagnostics for infectious diseases: From methods to devices

The current widespread of COVID-19 all over the world, which is caused by SARS-CoV-2 virus, has again emphasized the importance of development of point-of-care (POC) diagnostics for timely prevention and control of the pandemic. Compared with labor- and time-consuming traditional diagnostic methods, POC diagnostics exhibit several advantages such as faster diagnostic speed, better sensitivity and specificity, lower cost, higher efficiency and ability of on-site detection. To achieve POC diagnostics, developing POC detection methods and correlated POC devices is the key and should be given top priority. The fast development of microfluidics, micro electro-mechanical systems (MEMS) technology, nanotechnology and materials science, have benefited the production of a series of portable, miniaturized, low cost and highly integrated POC devices for POC diagnostics of various infectious diseases. In this review, various POC detection methods for the diagnosis of infectious diseases, including electrochemical biosensors, fluorescence biosensors, surface-enhanced Raman scattering (SERS)-based biosensors, colorimetric biosensors, chemiluminiscence biosensors, surface plasmon resonance (SPR)-based biosensors, and magnetic biosensors, were first summarized. Then, recent progresses in the development of POC devices including lab-on-a-chip (LOC) devices, lab-on-a-disc (LOAD) devices, microfluidic paper-based analytical devices (μPADs), lateral flow devices, miniaturized PCR devices, and isothermal nucleic acid amplification (INAA) devices, were systematically discussed. Finally, the challenges and future perspectives for the design and development of POC detection methods and correlated devices were presented. The ultimate goal of this review is to provide new insights and directions for the future development of POC diagnostics for the management of infectious diseases and contribute to the prevention and control of infectious pandemics like COVID-19.

Deep learning-enabled imaging flow cytometry for high-speed Cryptosporidium and Giardia detection

Imaging flow cytometry has become a popular technology for bioparticle image analysis because of its capability of capturing thousands of images per second. Nevertheless, the vast number of images generated by imaging flow cytometry imposes great challenges for data analysis especially when the species have similar morphologies. In this work, we report a deep learning-enabled high-throughput system for predicting Cryptosporidium and Giardia in drinking water. This system combines imaging flow cytometry and an efficient artificial neural network called MCellNet, which achieves a classification accuracy >99.6%. The system can detect Cryptosporidium and Giardia with a sensitivity of 97.37% and a specificity of 99.95%. The high-speed analysis reaches 346 frames per second, outperforming the state-of-the-art deep learning algorithm MobileNetV2 in speed (251 frames per second) with a comparable classification accuracy. The reported system empowers rapid, accurate, and high throughput bioparticle detection in clinical diagnostics, environmental monitoring and other potential biosensing applications.

Holotomography: Refractive Index as an Intrinsic Imaging Contrast for 3-D Label-Free Live Cell Imaging

Live cell imaging provides essential information in the investigation of cell biology and related pathophysiology. Refractive index (RI) can serve as intrinsic optical imaging contrast for 3-D label-free and quantitative live cell imaging, and provide invaluable information to understand various dynamics of cells and tissues for the study of numerous fields. Recently significant advances have been made in imaging methods and analysis approaches utilizing RI, which are now being transferred to biological and medical research fields, providing novel approaches to investigate the pathophysiology of cells. To provide insight into how RI can be used as an imaging contrast for imaging of biological specimens, here we provide the basic principle of RI-based imaging techniques and summarize recent progress on applications, ranging from microbiology, hematology, infectious diseases, hematology, and histopathology.

Single-cell cytometry via multiplexed fluorescence prediction by label-free reflectance microscopy

Traditional imaging cytometry uses fluorescence markers to identify specific structures but is limited in throughput by the labeling process. We develop a label-free technique that alleviates the physical staining and provides multiplexed readouts via a deep learning-augmented digital labeling method. We leverage the rich structural information and superior sensitivity in reflectance microscopy and show that digital labeling predicts accurate subcellular features after training on immunofluorescence images. We demonstrate up to three times improvement in the prediction accuracy over the state of the art. Beyond fluorescence prediction, we demonstrate that single cell-level structural phenotypes of cell cycles are correctly reproduced by the digital multiplexed images, including Golgi twins, Golgi haze during mitosis, and DNA synthesis. We further show that the multiplexed readouts enable accurate multiparametric single-cell profiling across a large cell population. Our method can markedly improve the throughput for imaging cytometry toward applications for phenotyping, pathology, and high-content screening.

High space-bandwidth in quantitative phase imaging using partially spatially coherent digital holographic microscopy and a deep neural network

Quantitative phase microscopy (QPM) is a label-free technique that enables monitoring of morphological changes at the subcellular level. The performance of the QPM system in terms of spatial sensitivity and resolution depends on the coherence properties of the light source and the numerical aperture (NA) of objective lenses. Here, we propose high space-bandwidth quantitative phase imaging using partially spatially coherent digital holographic microscopy (PSC-DHM) assisted with a deep neural network. The PSC source synthesized to improve the spatial sensitivity of the reconstructed phase map from the interferometric images. Further, compatible generative adversarial network (GAN) is used and trained with paired low-resolution (LR) and high-resolution (HR) datasets acquired from the PSC-DHM system. The training of the network is performed on two different types of samples, i.e. mostly homogenous human red blood cells (RBC), and on highly heterogeneous macrophages. The performance is evaluated by predicting the HR images from the datasets captured with a low NA lens and compared with the actual HR phase images. An improvement of 9× in the space-bandwidth product is demonstrated for both RBC and macrophages datasets. We believe that the PSC-DHM + GAN approach would be applicable in single-shot label free tissue imaging, disease classification and other high-resolution tomography applications by utilizing the longitudinal spatial coherence properties of the light source.

Calibration-free quantitative phase imaging using data-driven aberration modeling

We present a data-driven approach to compensate for optical aberrations in calibration-free quantitative phase imaging (QPI). Unlike existing methods that require additional measurements or a background region to correct aberrations, we exploit deep learning techniques to model the physics of aberration in an imaging system. We demonstrate the generation of a single-shot aberration-corrected field image by using a U-net-based deep neural network that learns a translation between an optical field with aberrations and an aberration-corrected field. The high fidelity and stability of our method is demonstrated on 2D and 3D QPI measurements of various confluent eukaryotic cells and microbeads, benchmarking against the conventional method using background subtractions.

Automatic detection and characterization of quantitative phase images of thalassemic red blood cells using a mask region-based convolutional neural network

SIGNIFICANCE

Label-free quantitative phase imaging is a promising technique for the automatic detection of abnormal red blood cells (RBCs) in real time. Although deep-learning techniques can accurately detect abnormal RBCs from quantitative phase images efficiently, their applications in diagnostic testing are limited by the lack of transparency. More interpretable results such as morphological and biochemical characteristics of individual RBCs are highly desirable.

AIM

An end-to-end deep-learning model was developed to efficiently discriminate thalassemic RBCs (tRBCs) from healthy RBCs (hRBCs) in quantitative phase images and segment RBCs for single-cell characterization.

APPROACH

Two-dimensional quantitative phase images of hRBCs and tRBCs were acquired using digital holographic microscopy. A mask region-based convolutional neural network (Mask R-CNN) model was trained to discriminate tRBCs and segment individual RBCs. Characterization of tRBCs was achieved utilizing SHapley Additive exPlanation analysis and canonical correlation analysis on automatically segmented RBC phase images.

RESULTS

The implemented model achieved 97.8% accuracy in detecting tRBCs. Phase-shift statistics showed the highest influence on the correct classification of tRBCs. Associations between the phase-shift features and three-dimensional morphological features were revealed.

CONCLUSIONS

The implemented Mask R-CNN model accurately identified tRBCs and segmented RBCs to provide single-RBC characterization, which has the potential to aid clinical decision-making.

Identification and Staging of B-Cell Acute Lymphoblastic Leukemia Using Quantitative Phase Imaging and Machine Learning

Identification and classification of leukemia cells in a rapid and label-free fashion is clinically challenging and thus presents a prime arena for implementing new diagnostic tools. Quantitative phase imaging, which maps optical path length delays introduced by the specimen, has been demonstrated to discern cellular phenotypes based on differential morphological attributes. Rapid acquisition capability and the availability of label-free images with high information content have enabled researchers to use machine learning (ML) to reveal latent features. We developed a set of ML classifiers, including convolutional neural networks, to discern healthy B cells from lymphoblasts and classify stages of B cell acute lymphoblastic leukemia. Here, we show that the average dry mass and volume of normal B cells are lower than those of cancerous cells and that these morphologic parameters increase further alongside disease progression. We find that the relaxed training requirements of a ML approach are conducive to the classification of cell type, with minimal space, training time, and memory requirements. Our findings pave the way for a larger study on clinical samples of acute lymphoblastic leukemia, with the overarching goal of its broader use in hematopathology, where the prospect of objective diagnoses with minimal sample preparation remains highly desirable.

Holography applications toward medical field: An overview

Purpose:

3D Holography is a commercially available, disruptive innovation, which can be customised as per the requirements and is supporting Industry 4.0. The purpose of this paper is to study the potential applications of 3D holography in the medical field. This paper explores the concept of holography and its significant benefits in the medical field.

Methods:

The paper is derived through the study of various research papers on Holography and its applications in the medical field. The study tries to identify the direction of research &development and see how this innovative technology can be used effectively for better treatment of patients.

Results:

Holography uses digital imaging inputs and provides an extensive visualisation of the data for training doctors, surgeons and students. Holography converts information about the body into a digital format and has the potential to inform, promote and entertain the medical students and doctors. However, it needs a large amount of space for data storage and extensive software support for analysis and skills for customising. This technology seems good to solve a variety of medical issues by storing and using patient data in developing 3D holograms, which are useful to assist successful treatment and surgery. It seems useful in providing flexible solutions in the area of medical research. Finally, the paper identifies 13 significant applications of this technology in the medical field and discusses them appropriately.

Conclusion:

The paper explores holographic applications in medical research due to its extensive capability of image processing. Holographic images are non-contact 3D images having a large field of depth. A physician can now zoom the holographic image for a better view of the medical part. This innovative technology can create advancements in the diagnosis and treatment process, which can improve medical practice. It helps in quick detection of problems in various organs like brain, heart, liver, kidney etc. By using this technology, medical practitioners can see colourful organs at multiple angles with better accuracy. It opens up an innovative way of planning, testing of procedures and diagnosis. With technological developments, compact hardware and software are now available to help medical research and related applications.

Classification of cell morphology with quantitative phase microscopy and machine learning

We describe and compare two machine learning approaches for cell classification based on label-free quantitative phase imaging with transport of intensity equation methods. In one approach, we design a multilevel integrated machine learning classifier including various individual models such as artificial neural network, extreme learning machine and generalized logistic regression. In another approach, we apply a pretrained convolutional neural network using transfer learning for the classification. As a validation, we show the performances of both approaches on classification between macrophages cultured in normal gravity and microgravity with quantitative phase imaging. The multilevel integrated classifier achieves average accuracy 93.1%, which is comparable to the average accuracy 93.5% obtained by convolutional neural network. The presented quantitative phase imaging system with two classification approaches could be helpful to biomedical scientists for easy and accurate cell analysis.

Deep learning-based cell identification and disease diagnosis using spatio-temporal cellular dynamics in compact digital holographic microscopy

We demonstrate a successful deep learning strategy for cell identification and disease diagnosis using spatio-temporal cell information recorded by a digital holographic microscopy system. Shearing digital holographic microscopy is employed using a low-cost, compact, field-portable and 3D-printed microscopy system to record video-rate data of live biological cells with nanometer sensitivity in terms of axial membrane fluctuations, then features are extracted from the reconstructed phase profiles of segmented cells at each time instance for classification. The time-varying data of each extracted feature is input into a recurrent bi-directional long short-term memory (Bi-LSTM) network which learns to classify cells based on their time-varying behavior. Our approach is presented for cell identification between the morphologically similar cases of cow and horse red blood cells. Furthermore, the proposed deep learning strategy is demonstrated as having improved performance over conventional machine learning approaches on a clinically relevant dataset of human red blood cells from healthy individuals and those with sickle cell disease. The results are presented at both the cell and patient levels. To the best of our knowledge, this is the first report of deep learning for spatio-temporal-based cell identification and disease detection using a digital holographic microscopy system.

Advancing theoretical understanding and practical performance of signal processing for nonlinear optical communications through machine learning

In long-haul optical communication systems, compensating nonlinear effects through digital signal processing (DSP) is difficult due to intractable interactions between Kerr nonlinearity, chromatic dispersion (CD) and amplified spontaneous emission (ASE) noise from inline amplifiers. Optimizing the standard digital back propagation (DBP) as a deep neural network (DNN) with interleaving linear and nonlinear operations for fiber nonlinearity compensation was shown to improve transmission performance in idealized simulation environments. Here, we extend such concepts to practical single-channel and polarization division multiplexed wavelength division multiplexed experiments. We show improved performance compared to state-of-the-art DSP algorithms and additionally, the optimized DNN-based DBP parameters exhibit a mathematical structure which guides us to further analyze the noise statistics of fiber nonlinearity compensation. This machine learning-inspired analysis reveals that ASE noise and incomplete CD compensation of the Kerr nonlinear term produce extra distortions that accumulates along the DBP stages. Therefore, the best DSP should balance between suppressing these distortions and inverting the fiber propagation effects, and such trade-off shifts across different DBP stages in a quantifiable manner. Instead of the common 'black-box' approach to intractable problems, our work shows how machine learning can be a complementary tool to human analytical thinking and help advance theoretical understandings in disciplines such as optics.