Series Machine Learning Research Articles

ERGO-ML: comparing IllustrisTNG and HSC galaxy images via contrastive learning

ABSTRACT Modern cosmological hydrodynamical galaxy simulations provide tens of thousands of reasonably realistic synthetic galaxies across cosmic time. However, quantitatively assessing the level of realism of simulated universes in comparison to the real one is difficult. In this paper of the Extracting Reality from Galaxy Observables with Machine Learning series, we utilize contrastive learning to directly compare a large sample of simulated and observed galaxies based on their stellar-light images. This eliminates the need to specify summary statistics and allows to exploit the whole information content of the observations. We produce survey-realistic galaxy mock data sets resembling real Hyper Suprime-Cam (HSC) observations using the cosmological simulations TNG50 and TNG100. Our focus is on galaxies with stellar masses between 109 and 1012 M⊙ at z = 0.1–0.4. This allows us to evaluate the realism of the simulated TNG galaxies in comparison to actual HSC observations. We apply the self-supervised contrastive learning method Nearest Neighbour Contrastive Learning to the images from both simulated and observed data sets (g-, r-, i-bands). This results in a 256-dimensional representation space, encoding all relevant observable galaxy properties. First, this allows us to identify simulated galaxies that closely resemble real ones by seeking similar images in this multidimensional space. Even more powerful, we quantify the alignment between the representations of these two image sets, finding that the majority (≳ 70 per cent) of the TNG galaxies align well with observed HSC images. However, a subset of simulated galaxies with larger sizes, steeper Sérsic profiles, smaller Sérsic ellipticities, and larger asymmetries appears unrealistic. We also demonstrate the utility of our derived image representations by inferring properties of real HSC galaxies using simulated TNG galaxies as the ground truth.

Lanthanide-Doped ZnO Nanoparticles: Unraveling Their Role in Cytotoxicity, Antioxidant Capacity, and Nanotoxicology.

This study used a sonochemical synthesis method to prepare (La, Sm)-doped ZnO nanoparticles (NPs). The effect of incorporating these lanthanide elements on the structural, optical, and morphological properties of ZnO-NPs was analyzed. The cytotoxicity and the reactive oxygen species (ROS) generation capacity of ZnO-NPs were evaluated against breast (MCF7) and colon (HT29) cancer cell lines. Their antioxidant activity was analyzed using a DPPH assay, and their toxicity towards Artemia salina nauplii was also evaluated. The results revealed that treatment with NPs resulted in the death of 10.559-42.546% and 18.230-38.643% of MCF7 and HT29 cells, respectively. This effect was attributed to the ability of NPs to downregulate ROS formation within the two cell lines in a dose-dependent manner. In the DPPH assay, treatment with (La, Sm)-doped ZnO-NPs inhibited the generation of free radicals at IC50 values ranging from 3.898 to 126.948 μg/mL. Against A. salina nauplii, the synthesized NPs did not cause death nor induce morphological changes at the tested concentrations. A series of machine learning (ML) models were used to predict the biological performance of (La, Sm)-doped ZnO-NPs. Among the designed ML models, the gradient boosting model resulted in the greatest mean absolute error (MAE) (MAE 9.027, R2 = 0.86). The data generated in this work provide innovative insights into the influence of La and Sm on the structural arrangement and chemical features of ZnO-NPs, together with their cytotoxicity, antioxidant activity, and in vivo toxicity.

Machine learning in healthcare strategic management: a systematic literature review

PurposeThis study specifically seeks to investigate the strategic implementation of machine learning (ML) algorithms and techniques in healthcare institutions to enhance innovation management in healthcare settings.Design/methodology/approach The papers from 2011 to 2021 were considered following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. First, relevant keywords were identified, and screening was performed. Bibliometric analysis was performed. One hundred twenty-three relevant documents that passed the eligibility criteria were finalized.Findings Overall, the annual scientific production section results reveal that ML in the healthcare sector is growing significantly. Performing bibliometric analysis has helped find unexplored areas; understand the trend of scientific publication; and categorize topics based on emerging, trending and essential. The paper discovers the influential authors, sources, countries and ML and healthcare management keywords.Research limitations/implications The study helps understand various applications of ML in healthcare institutions, such as the use of Internet of Things in healthcare, the prediction of disease, finding the seriousness of a case, natural language processing, speech and language-based classification, etc. This analysis would help future researchers and developers target the healthcare sector areas that are likely to grow in the coming future.Practical implications The study highlights the potential for ML to enhance medical support within healthcare institutions. It suggests that regression algorithms are particularly promising for this purpose. Hospital management can leverage time series ML algorithms to estimate the number of incoming patients, thus increasing hospital availability and optimizing resource allocation. ML has been instrumental in the development of these systems. By embracing telemedicine and remote monitoring, healthcare management can facilitate the creation of online patient surveillance and monitoring systems, allowing for early medical intervention and ultimately improving the efficiency and effectiveness of medical services.Originality/value By offering a comprehensive panorama of ML's integration within healthcare institutions, this study underscores the pivotal role of innovation management in healthcare. The findings contribute to a holistic understanding of ML's applications in healthcare and emphasize their potential to transform and optimize healthcare delivery.

Prediction of Cooling Load of Tropical Buildings with Machine Learning

Cooling load refers to the amount of energy to be removed from a space (or consumed) to bring that space to an acceptable temperature or to maintain the temperature of a space at an acceptable range. The study aimed to develop a series of models and determine the most accurate ones in the prediction of the cooling load of low-rise tropical buildings based on their basic architectural and structural characteristics. In this context, a series of machine learning (regression) algorithms were tested during the research to determine the most accurate/efficient prediction model. In this regard, a data set consisting of ten features indicating the basic characteristics of the building (floor area, aspect ratio, ceiling height, window material, external wall material, roof material, window wall ratio north faced, window wall ratio south faced, horizontal shading, orientation) were used to predict the cooling load of a low-rise tropical building. The dataset was generated utilizing a set of generative and algorithmic design tools. Following the dataset generation, a series of regression models were tested to find the most accurate model to predict the cooling load. The results of the tests with different algorithms revealed that the relationship between the predictor variables and cooling load could be efficiently modeled through Histogram Gradient Boosting and Stacking models.

Clinical Significance and Immune Infiltration Analyses of the Cuproptosis-Related Human Copper Proteome in Gastric Cancer.

Background: The human copper Cu proteome, also termed Cu-binding proteins (CBP), is responsible for transporting “free” Cu to the cell that is related to cuproptosis. However, their role in gastric cancer (GC) has not been reported. Methods: RNA expression data of 946 GC patients were collected. A series of machine learning and bioinformatic approaches were combined to build a CBP signature to predict survival and immune microenvironment and guide the priority treatment. Immunohistochemistry and multicolor immunofluorescence (mIF) in 1076 resection slides were used to verify immune features. Results: A CBP signature was constructed using the machine learning method from TCGA that classifies cases as CBP_low and CBP_high groups. Multivariable Cox analysis confirmed that the CBP signature was an independent prognostic factor in the training and validation cohorts. Additionally, GC patients with low CBPscores showed an increase in anti-tumor immune microenvironment, which was further verified by mIF in pathological resections following immunotherapy. Importantly, patients with low CBPscores had higher levels of TMB/MSI and responded well to immunotherapy. Conclusions: We conducted the first and comprehensive CBP analysis of GC patients and established a clinically feasible CBP signature for predicting survival and response to treatment, which will be helpful for guiding personalized medicine.

Multi‐criteria approach for handling sophisticated data transmission over gateways in blockchain and Internet of Things (IoT) federated networks

AbstractIn the real‐world scenario, the Internet of Things (IoT) and blockchain federated networks implicate smart devices for gathering the data and then directing the data to the destined nodes through gateway devices. A gateway usually supports a number of wireless sensor networks (WSNs) irrespective of their underlying communication technologies. Furthermore, a gateway usually supports a number of WSNs with blockchain master nodes, which employ diverse communication technologies. As a result, gateways are meant to play a versatile role without worrying about the underlying architecture, hardware, and software details of the connecting nodes in blockchain federated networks. This paper proposes a two‐phased methodology for blockchain and IoT federated networks where efficient link selection based on a multi‐criteria‐based approach is performed. The dynamic gateway scheduling strategy is capable enough to support blockchain‐based transactions as well as communication in IoT devices. Furthermore, the proposed methodology enhances data transfer fairness for each gateway, resulting in efficient data transmission. Machine learning (ML) methods are also devised to analyze the status of the communication channels before employing the link selection mechanism. Then the selection of links is performed based on multi‐criteria statistical techniques. Finally, the scheduling is performed for selecting the appropriate gateway for channelizing the blockchain data speedily. The selection of links is made based on the dynamic status of the links. The usage of time series ML methods such as LSTM is made to predict the upcoming traffic on the links. Multiple criteria‐based statistical approaches are utilized for the selection of the link optimally. Finally, the scheduling is performed to fulfil the criteria of 6G networks, where network users are rewarded with seamless connectivity and uninterrupted services. The efficacy of the proposed two‐phased mechanism is demonstrated through simulation results. The results are demonstrated with respect to the total data conveyed, packet delivery ratio (PDR), energy consumed by the links, and throughput of the network. The PDR obtained by the proposed work for the high volume of data is 93% and for the low volume of data is 97%. The packet delay is reduced by 21% for low volume data and 19% for high volume data by the proposed methods. The data transmission is also optimized remarkably by the proposed method. The proposed fairness‐driven mechanism (FDM) is compared with the state‐of‐the‐art methods to prove the viability and robustness of the two‐phased proposed mechanism of link selection for data transmission over blockchain and IoT federated networks.

Focal Adhesion-Related Signatures Predict the Treatment Efficacy of Chemotherapy and Prognosis in Patients with Gastric Cancer.

BackgroundThe current tumor-node-metastasis (TNM) staging system is insufficient for predicting the efficacy of chemotherapy in patients with gastric cancer (GC). This study aimed to analyze the association between the focal adhesion pathway and therapeutic efficacy of chemotherapy in patients with GC.MethodsRNA sequencing was performed on 33 clinical samples from patients who responded or did not respond to treatment prior to neoadjuvant chemotherapy. The validation sets containing 696 GC patients with RNA data from three cohorts (PKUCH, TCGA, and GSE14210) were analyzed. A series of machine learning and bioinformatics approaches was combined to build a focal adhesion-related signature model to predict the treatment efficacy and prognosis of patients with GC.ResultsAmong the various signaling pathways associated with cancer, focal adhesion was identified as a risk factor related to the treatment efficacy of chemotherapy and prognosis in patients with GC. The focal adhesion-related gene model (FAscore) discriminated patients with a high FAscore who are insensitive to neoadjuvant chemotherapy in our training cohort, and the predicted value was further verified in the GSE14210 cohort. Survival analysis also demonstrated that patients with high FAscores had a relatively shorter survival compared to those with low FAscores. In addition, we found that the levels of tumor mutation burden (TMB) and microsatellite instability (MSI) increased with an increase in FAscore, and the tumor microenvironment (TME) also shifted to a pro-tumor immune microenvironment.ConclusionThe FAscore model can be used to predict the treatment efficacy of chemotherapy and select appropriate treatment strategies for patients with GC.

CfTrack: A Method of Exome-Wide Mutation Analysis of Cell-free DNA to Simultaneously Monitor the Full Spectrum of Cancer Treatment Outcomes Including MRD, Recurrence, and Evolution.

Cell-free DNA (cfDNA) offers a noninvasive approach to monitor cancer. Here we develop a method using whole-exome sequencing (WES) of cfDNA for simultaneously monitoring the full spectrum of cancer treatment outcomes, including minimal residual disease (MRD), recurrence, evolution, and second primary cancers. Three simulation datasets were generated from 26 patients with cancer to benchmark the detection performance of MRD/recurrence and second primary cancers. For further validation, cfDNA samples (n = 76) from patients with cancer (n = 35) with six different cancer types were used for performance validation during various treatments. We present a cfDNA-based cancer monitoring method, named cfTrack. Taking advantage of the broad genome coverage of WES data, cfTrack can sensitively detect MRD and cancer recurrence by integrating signals across known clonal tumor mutations of a patient. In addition, cfTrack detects tumor evolution and second primary cancers by de novo identifying emerging tumor mutations. A series of machine learning and statistical denoising techniques are applied to enhance the detection power. On the simulation data, cfTrack achieved an average AUC of 99% on the validation dataset and 100% on the independent dataset in detecting recurrence in samples with tumor fractions ≥0.05%. In addition, cfTrack yielded an average AUC of 88% in detecting second primary cancers in samples with tumor fractions ≥0.2%. On real data, cfTrack accurately monitors tumor evolution during treatment, which cannot be accomplished by previous methods. Our results demonstrated that cfTrack can sensitively and specifically monitor the full spectrum of cancer treatment outcomes using exome-wide mutation analysis of cfDNA.

Secondary porosity prediction in complex carbonate reefs using 3D CT scan image analysis and machine learning

Development and preservation of carbonate porosity and permeability are critical to characterizing reservoirs. However, secondary porosity, such as vugs and fractures, are difficult to identify and require collection of expensive wireline tools or core sampling. Wireline logs and cores have traditionally been used to identify the presence of secondary porosity but fail to quantify the contribution to total reservoir porosity. Additionally, advanced wireline logs and core are not readily available for most wells. Dual energy CT scans were collected on whole core from the A-1 Carbonate and Brown Niagaran formations drawn from six wells in northern Michigan. 3D analysis techniques were applied to identify and isolate secondary porosity features. A series of machine learning and data analytics techniques were applied to the dataset to predict secondary porosity features on basic wireline logs. The predictive model successfully predicted secondary porosity with high confidence.

Extracting Data at Scale: Machine learning at the Natural History Museum

The Natural History Museum holds over 80 million specimens and 300 million pages of scientific text. This information is a vital research tool to help solve the most important challenge humans face over the coming years – mapping a sustainable future for ourselves and the ecosystems on which we depend. Digitising these collections and providing the data in a structured, computable form is a mammoth challenge. As of 2020, less than 15% of available specimen information currently residing on specimen labels or physical registers is digitised and publicly available (Walton et al. 2020). Machine learning applications can deliver a step-change in our activities’ scope, scale, and speed (Borsch et al. 2020). As part of SYNTHESYS+, the Natural History Museum is leading on the development of a cloud-based workflow platform for natural science specimens, the Specimen Data Refinery (SDR) (Smith et al. 2019). The SDR will provide a series of Machine Learning (ML) models, ranging from semantic segmentation to identify regions of interest on labels, to natural language processing to extract locality and taxonomic text entities from the labels, and image analysis to identify specimen traits and collection quality metrics. Each ML task is atomic, with users of the SDR selecting which model would best extract data from their digitised specimen images, allowing the workflows to be used in different institutions worldwide. It also solves one of the key problems in developing ML-based applications: the rapidity at which models become obsolete. New ML models can be introduced into the workflow, with incremental changes to improve processing, without interruption or refactoring of the pipeline. Alongside specimens, digitised images of pages of scientific literature provide another vital source of data. Functional traits mediate the interactions between plant species and their environment and play roles in determining species’ range size and threatened status. Such information is contained within the taxonomic descriptions of species and a natural language processing library has been developed to locate and extract plant functional traits from these texts (Hoehndorf et al. 2016). The ML models allow complex interrelationships between taxa and trait entities to be inferred based on the grammatical structure of sentences, improving the accuracy and extent of data point extraction. These two projects, like many other applications of ML in natural history collections, are focused on the extraction of visible information, for example, a piece of text or a measurable trait. Given the image of the specimen or page, a person would be able to extract the self-same information. However, ML excels in pattern matching and inferring unknown characters from an entire corpus. At the museum, we have started exploring this space, with our voyagerAI project for identifying specimens collected on historical expeditions of scientific discovery (e.g., the voyages of the Beagle and Challenger). This process fills in the gaps in specimen provenance and identifies 'lost' specimens collected by some of the most famous names in biodiversity history. Developing new applications of ML to uncover scientific meaning and tell the narratives of our collections, will be at the forefront of our scientific innovation in the coming years. This presentation will give an overview of these projects, and our future plans for using ML to extract data at scale within the Natural History Museum.

Machine Learning in Nuclear Medicine: Part 2-Neural Networks and Clinical Aspects.

This article is the second part in our machine learning series. Part 1 provided a general overview of machine learning in nuclear medicine. Part 2 focuses on neural networks. We start with an example illustrating how neural networks work and a discussion of potential applications. Recognizing that there is a spectrum of applications, we focus on recent publications in the areas of image reconstruction, low-dose PET, disease detection, and models used for diagnosis and outcome prediction. Finally, since the way machine learning algorithms are reported in the literature is extremely variable, we conclude with a call to arms regarding the need for standardized reporting of design and outcome metrics and we propose a basic checklist our community might follow going forward.