Tissue post-classification using the measured acoustic signals during 355 nm laser atherectomy procedures

The current laser atherectomy technologies to treat patients with challenging (to-cross) total chronic occlusions with a step-by-step (SBS) approach (without leading guide wire), are lacking real-time signal monitoring of the ablated tissues, and carry the risk for vessel perforation. We present first time post-classification of ablated tissues using acoustic signals recorded by a microphone placed nearby during five atherectomy procedures using 355 nm solid-state Auryon laser device performed with an SBS approach, some with highly severe calcification. Using our machine-learning algorithm, the classification results of these ablation signals recordings from five patients showed 93.7% classification accuracy with arterial vs nonarterial wall material. While still very preliminary and requiring a larger study and thereafter as commercial device, the results of these first acoustic post-classification in SBS cases are very promising. This study implies, as a general statement, that online recording of the acoustic signals using a noncontact microphone, may potentially serve for an online classification of the ablated tissue in SBS cases. This technology could be used to confirm correct positioning in the vasculature, and by this, to potentially further reduce the risk of perforation using 355 nm laser atherectomy in such procedures.     

In situ tissue classification during laser ablation using acoustic signals

We suggest a novel method to classify the type of tissue that is being ablated, using the recorded acoustic sound waves during pulsed ultraviolet laser ablation. The motivation of the current research is tissue classification during vascular interventions, where the identification of the ablated tissue is vital. We classify the acoustic signatures using Mel‐frequency cepstral coefficients (MFCCs) feature extraction with a Support Vector Machine (SVM) algorithm, and in addition, use a fully connected deep neural network (FC‐DNN) algorithm. First, we classify three different liquids using our method as a preliminary proof of concept. Then, we classify ex vivo porcine aorta and bovine tendon tissues in the presence of saline. Finally, we classify ex vivo porcine aorta and bovine tendon tissues where the acoustic signals are recorded through chicken breast medium. The results for tissue classification in saline and through chicken breast both show high accuracy (>98%), based on tens of thousands of acoustic signals for each experiment. The experiments were conducted in a noisy and challenging setting that tries to imitate practical working conditions. The obtained results could pave the way towards practical tissue classification in various important medical procedures, achieving enhanced efficacy and safety.     

Machine learning techniques for protein function prediction

Proteins play important roles in living organisms, and their function is directly linked with their structure. Due to the growing gap between the number of proteins being discovered and their functional characterization (in particular as a result of experimental limitations), reliable prediction of protein function through computational means has become crucial. This paper reviews the machine learning techniques used in the literature, following their evolution from simple algorithms such as logistic regression to more advanced methods like support vector machines and modern deep neural networks. Hyperparameter optimization methods adopted to boost prediction performance are presented. In parallel, the metamorphosis in the features used by these algorithms from classical physicochemical properties and amino acid composition, up to text-derived features from biomedical literature and learned feature representations using autoencoders, together with feature selection and dimensionality reduction techniques, are also reviewed. The success stories in the application of these techniques to both general and specific protein function prediction are discussed.     

面向蛋白质功能位点识别的机器学习平台构建

有关蛋白质功能的研究是解析生命奥秘的基础,机器学习技术在该领域已有广泛应用。利用支持向量机(support vectormachine,SVM)方法,构建一个预测蛋白质功能位点的通用平台。该平台先提取非同源蛋白质序列,再对这些序列进行特征编码(包括序列的基本信息、物化特征、结构信息及序列保守性特征等),以编码好的样本作为训练数据,利用SVM进行训练,得到敏感性、特异性、Matthew相关系数、准确率及ROC曲线等评价指标,反复测试,得到评价指标最优的SVM模型后,便可以用来预测蛋白质序列上的功能位点。该平台除了应用在预测蛋白质功能位点之外,还可以应用于疾病相关单核苷酸多态性(SNP)预测分析、预测蛋白质结构域分析、生物分子间的相互作用等。     

Machine learning for epigenetics and future medical applications

Understanding epigenetic processes holds immense promise for medical applications. Advances in Machine Learning (ML) are critical to realize this promise. Previous studies used epigenetic data sets associated with the germline transmission of epigenetic transgenerational inheritance of disease and novel ML approaches to predict genome-wide locations of critical epimutations. A combination of Active Learning (ACL) and Imbalanced Class Learning (ICL) was used to address past problems with ML to develop a more efficient feature selection process and address the imbalance problem in all genomic data sets. The power of this novel ML approach and our ability to predict epigenetic phenomena and associated disease is suggested. The current approach requires extensive computation of features over the genome. A promising new approach is to introduce Deep Learning (DL) for the generation and simultaneous computation of novel genomic features tuned to the classification task. This approach can be used with any genomic or biological data set applied to medicine. The application of molecular epigenetic data in advanced machine learning analysis to medicine is the focus of this review.     

Machine learning approaches for prediction of linear B-cell epitopes on proteins

Identification and characterization of antigenic determinants on proteins has received considerable attention utilizing both, experimental as well as computational methods. For computational routines mostly structural as well as physicochemical parameters have been utilized for predicting the antigenic propensity of protein sites. However, the performance of computational routines has been low when compared to experimental alternatives. Here we describe the construction of machine learning based classifiers to enhance the prediction quality for identifying linear B-cell epitopes on proteins. Our approach combines several parameters previously associated with antigenicity, and includes novel parameters based on frequencies of amino acids and amino acid neighborhood propensities. We utilized machine learning algorithms for deriving antigenicity classification functions assigning antigenic propensities to each amino acid of a given protein sequence. We compared the prediction quality of the novel classifiers with respect to established routines for epitope scoring, and tested prediction accuracy on experimental data available for HIV proteins. The major finding is that machine learning classifiers clearly outperform the reference classification systems on the HIV epitope validation set.     

Real‐time monitoring of incision profile during laser surgery using shock wave detection

Lack of sensory feedback during laser surgery prevents surgeons from discerning the exact location of the incision, which increases duration and complexity of the treatment. In this study we demonstrate a new method for monitoring of laser ablation procedures. Real‐time tracking of the exact three dimensional (3D) lesion profile is accomplished by detection of shock waves emanating from the ablation spot and subsequent reconstruction of the incision location using time‐of‐flight data obtained from multiple acoustic detectors. Here, incisions of up to 9 mm in depth, created by pulsed laser ablation of fresh bovine tissue samples, were successfully monitored in real time. It was further observed that, by utilizing as little as 12 detection elements, the incision profile can be characterized with accuracy below 0.5 mm in all three dimensions and in good agreement with histological examinations. The proposed method holds therefore promise for delivering high precision real‐time feedback during laser surgeries. (© 2013 WILEY‐VCH Verlag GmbH &Co. KGaA, Weinheim)     

Machine learning for accelerating process-based computation of land biogeochemical cycles

Global change ecology nowadays embraces ever-growing large observational datasets (big-data) and complex mathematical models that track hundreds of ecological processes (big-model). The rapid advancement of the big-data-big-model has reached its bottleneck: high computational requirements prevent further development of models that need to be integrated over long time-scales to simulate the distribution of ecosystems carbon and nutrient pools and fluxes. Here, we introduce a machine-learning acceleration (MLA) tool to tackle this grand challenge. We focus on the most resource-consuming step in terrestrial biosphere models (TBMs): the equilibration of biogeochemical cycles (spin-up), a prerequisite that can take up to 98% of the computational time. Through three members of the ORCHIDEE TBM family part of the IPSL Earth System Model, including versions that describe the complex interactions between nitrogen, phosphorus and carbon that do not have any analytical solution for the spin-up, we show that an unoptimized MLA reduced the computation demand by 77%–80% for global studies via interpolating the equilibrated state of biogeochemical variables for a subset of model pixels. Despite small biases in the MLA-derived equilibrium, the resulting impact on the predicted regional carbon balance over recent decades is minor. We expect a one-order of magnitude lower computation demand by optimizing the choices of machine learning algorithms, their settings, and balancing the trade-off between quality of MLA predictions and need for TBM simulations for training data generation and bias reduction. Our tool is agnostic to gridded models (beyond TBMs), compatible with existing spin-up acceleration procedures, and opens the door to a wide variety of future applications, with complex non-linear models benefit most from the computational efficiency.     

Machine learning in optical coherence tomography angiography

Optical coherence tomography angiography (OCTA) offers a noninvasive label-free solution for imaging retinal vasculatures at the capillary level resolution. In principle, improved resolution implies a better chance to reveal subtle microvascular distortions associated with eye diseases that are asymptomatic in early stages. However, massive screening requires experienced clinicians to manually examine retinal images, which may result in human error and hinder objective screening. Recently, quantitative OCTA features have been developed to standardize and document retinal vascular changes. The feasibility of using quantitative OCTA features for machine learning classification of different retinopathies has been demonstrated. Deep learning-based applications have also been explored for automatic OCTA image analysis and disease classification. In this article, we summarize recent developments of quantitative OCTA features, machine learning image analysis, and classification.     

Machine learning approaches in MALDI-MSI: clinical applications

Introduction: Despite the unquestionable advantages of Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Imaging in visualizing the spatial distribution and the relative abundance of biomolecules directly on-tissue, the yielded data is complex and high dimensional. Therefore, analysis and interpretation of this huge amount of information is mathematically, statistically and computationally challenging.

Areas covered: This article reviews some of the challenges in data elaboration with particular emphasis on machine learning techniques employed in clinical applications, and can be useful in general as an entry point for those who want to study the computational aspects. Several characteristics of data processing are described, enlightening advantages and disadvantages. Different approaches for data elaboration focused on clinical applications are also provided. Practical tutorial based upon Orange Canvas and Weka software is included, helping familiarization with the data processing.

Expert commentary: Recently, MALDI-MSI has gained considerable attention and has been employed for research and diagnostic purposes, with successful results. Data dimensionality constitutes an important issue and statistical methods for information-preserving data reduction represent one of the most challenging aspects. The most common data reduction methods are characterized by collecting independent observations into a single table. However, the incorporation of relational information can improve the discriminatory capability of the data.     

Femtosecond pulsed laser ablation to enhance drug delivery across the skin

Laser poration of the skin locally removes its outermost, barrier layer, and thereby provides a route for the diffusion of topically applied drugs. Ideally, no thermal damage would surround the pores created in the skin, as tissue coagulation would be expected to limit drug diffusion. Here, a femtosecond pulsed fiber laser is used to porate mammalian skin ex vivo. This first application of a hollow core negative curvature fiber (HC‐NCF) to convey a femtosecond pulsed, visible laser beam results in reproducible skin poration. The effect of applying ink to the skin surface, prior to ultra‐short pulsed ablation, has been examined and Raman spectroscopy reveals that the least, collateral thermal damage occurs in inked skin. Pre‐application of ink reduces the laser power threshold for poration, an effect attributed to the initiation of plasma formation by thermionic electron emission from the dye in the ink. Poration under these conditions significantly increases the percutaneous permeation of caffeine in vitro. Dye‐enhanced, plasma‐mediated ablation of the skin is therefore a potentially advantageous approach to enhance topical/transdermal drug absorption. The combination of a fiber laser and a HC‐NCF, capable of emitting and delivering femtosecond pulsed, visible light, may permit a compact poration device to be developed.

Using a femtosecond pulsed, visible laser beam to create an array of micropores in dyed mammalian skin, with little collateral, thermal damage, leads to an enhancement in the percutaneous permeation of caffeine in vitro.     

Machine learning based tissue analysis reveals Brachyury has a diagnosis value in breast cancer

Background: The aim of the present study was to confirm the role of Brachyury in breast cancer and to verify whether four types of machine learning models can use Brachyury expression to predict the survival of patients.Methods: We conducted a retrospective review of the medical records to obtain patient information, and made the patient’s paraffin tissue into tissue chips for staining analysis. We selected 303 patients for research and implemented four machine learning algorithms, including multivariate logistic regression model, decision tree, artificial neural network and random forest, and compared the results of these models with each other. Area under the receiver operating characteristic (ROC) curve (AUC) was used to compare the results.Results: The chi-square test results of relevant data suggested that the expression of Brachyury protein in cancer tissues was significantly higher than that in paracancerous tissues (P=0.0335); patients with breast cancer with high Brachyury expression had a worse overall survival (OS) compared with patients with low Brachyury expression. We also found that Brachyury expression was associated with ER expression (P=0.0489). Subsequently, we used four machine learning models to verify the relationship between Brachyury expression and the survival of patients with breast cancer. The results showed that the decision tree model had the best performance (AUC = 0.781).Conclusions: Brachyury is highly expressed in breast cancer and indicates that patients had a poor prognosis. Compared with conventional statistical methods, decision tree model shows superior performance in predicting the survival status of patients with breast cancer.     

Integrated effects of fractional laser microablation and sonophoresis on skin immersion optical clearing in vivo

This study is aimed to find an approach for effective skin optical clearing in vivo using polyethylene glycol 300 (PEG‐300) as an optical clearing agent in combination with physical enhancers: fractional laser microablation (FLMA) and/or low‐frequency sonophoresis. In this study albino outbred rats were used. Light attenuation coefficient and optical clearing potential (OCP) of these approaches were evaluated in upper (from ~70 to ~200 μm) and middle (from ~200 to ~400 μm) dermis separately using optical coherence tomography. In 30 minutes, OCP of sonophoresis in combination with FLMA and PEG‐300 in the upper dermis was the maximal (2.3 ± 0.4) in comparison with other treatments in this time point. The most effective approach for optical clearing of middle dermis was PEG‐300 and sonophoresis; but the maximal value of OCP (1.6 ± 0.1) was achieved only in 90 minutes.     

Deep learning for computational biology

Technological advances in genomics and imaging have led to an explosion of molecular and cellular profiling data from large numbers of samples. This rapid increase in biological data dimension and acquisition rate is challenging conventional analysis strategies. Modern machine learning methods, such as deep learning, promise to leverage very large data sets for finding hidden structure within them, and for making accurate predictions. In this review, we discuss applications of this new breed of analysis approaches in regulatory genomics and cellular imaging. We provide background of what deep learning is, and the settings in which it can be successfully applied to derive biological insights. In addition to presenting specific applications and providing tips for practical use, we also highlight possible pitfalls and limitations to guide computational biologists when and how to make the most use of this new technology.     

Micropatterning and characterization of electrospun poly(ε‐caprolactone)/gelatin nanofiber tissue scaffolds by femtosecond laser ablation for tissue engineering applications

Experimental investigations aimed at assessing the effectiveness of femtosecond (FS) laser ablation for creating microscale features on electrospun poly(ε‐caprolactone) (PCL)/gelatin nanofiber tissue scaffold capable of controlling cell distribution are described. Statistical comparisons of the fiber diameter and surface porosity on laser‐machined and as‐spun surface were made and results showed that laser ablation did not change the fiber surface morphology. The minimum feature size that could be created on electrospun nanofiber surfaces by direct‐write ablation was measured over a range of laser pulse energies. The minimum feature size that could be created was limited only by the pore size of the scaffold surface. The chemical states of PCL/gelatin nanofiber surfaces were measured before and after FS laser machining by attenuated total reflectance Fourier transform infrared (ATR‐FTIR) spectroscopy and X‐ray photoelectron spectroscopy (XPS) and showed that laser machining produced no changes in the chemistry of the surface. In vitro, mouse embryonic stem cells (mES cells) were cultured on as‐spun surfaces and in laser‐machined microwells. Cell densities were found to be statistically indistinguishable after 1 and 2 days of growth. Additionally, confocal microscope imaging confirmed that spreading of mES cells cultured within laser‐machined microwells was constrained by the cavity walls, the expected and desired function of these cavities. The geometric constraint caused statistically significant smaller density of cells in microwells after 3 days of growth. It was concluded that FS laser ablation is an effective process for microscale structuring of these electrospun nanofiber tissue scaffold surfaces. Biotechnol. Bioeng. 2011; 108:116–126. © 2010 Wiley Periodicals, Inc.     

Machine Learning Methods for Exploring Sequence Determinants of 3D Genome Organization

In higher eukaryotic cells, chromosomes are folded inside the nucleus. Recent advances in whole-genome mapping technologies have revealed the multiscale features of 3D genome organization that are intertwined with fundamental genome functions. However, DNA sequence determinants that modulate the formation of 3D genome organization remain poorly characterized. In the past few years, predicting 3D genome organization based on DNA sequence features has become an active area of research. Here, we review the recent progress in computational approaches to unraveling important sequence elements for 3D genome organization. In particular, we discuss the rapid development of machine learning-based methods that facilitate the connections between DNA sequence features and 3D genome architectures at different scales. While much progress has been made in developing predictive models for revealing important sequence features for 3D genome organization, new research is urgently needed to incorporate multi-omic data and enhance model interpretability, further advancing our understanding of gene regulation mechanisms through the lens of 3D genome organization.     

Recruiting machine learning methods for molecular simulations of proteins

Abstract

Molecular dynamics (MD) simulations are critical to understanding the movements of proteins in time. Yet, MD simulations are limited due to the availability of high-resolution protein structures, accuracy of the underlying force-field, computational expense, and difficulty in analysing big data-sets. Machine learning algorithms are now routinely used to circumvent many of these limitations and computational biophysicists are continuously making progress in developing novel applications. Here, we discuss some of these methods, varying from traditional dimensionality reduction approaches to more recent abstractions such as transfer learning and reinforcement learning, and how they have been used to deal with the challenges in MD. We conclude with the prospective issues in the application of machine learning methods in MD, to increase accuracy and efficiency of protein dynamics studies in general.     

Machine learning for target discovery in drug development

The discovery of macromolecular targets for bioactive agents is currently a bottleneck for the informed design of chemical probes and drug leads. Typically, activity profiling against genetically manipulated cell lines or chemical proteomics is pursued to shed light on their biology and deconvolute drug–target networks. By taking advantage of the ever-growing wealth of publicly available bioactivity data, learning algorithms now provide an attractive means to generate statistically motivated research hypotheses and thereby prioritize biochemical screens. Here, we highlight recent successes in machine intelligence for target identification and discuss challenges and opportunities for drug discovery.