Improving the Quality of Care in Radiation Oncology using Artificial Intelligence

Radiation therapy is a complex process involving multiple professionals and steps from simulation to treatment planning to delivery, and these procedures are prone to error. Additionally, the imaging and treatment delivery equipment in radiotherapy is highly complex and interconnected and represents another risk point in the quality of care. Numerous quality assurance tasks are carried out to ensure quality and to detect and prevent potential errors in the process of care. Recent developments in artificial intelligence provide potential tools to the radiation oncology community to improve the efficiency and performance of quality assurance efforts. Targets for artificial intelligence enhancement include the quality assurance of treatment plans, target and tissue structure delineation used in the plans, delivery of the plans and the radiotherapy delivery equipment itself. Here we review recent developments of artificial intelligence applications that aim to improve quality assurance processes in radiation therapy and discuss some of the challenges and limitations that require further development work to realise the potential of artificial intelligence for quality assurance.

Delta radiomics: a systematic review

BACKGROUND

Radiomics can provide quantitative features from medical imaging that can be correlated with various biological features and clinical endpoints. Delta radiomics, on the other hand, consists in the analysis of feature variation at different acquisition time points, usually before and after therapy. The aim of this study was to provide a systematic review of the different delta radiomics approaches.

METHODS

Eligible articles were searched in Embase, PubMed, and ScienceDirect using a search string that included free text and/or Medical Subject Headings (MeSH) with three key search terms: "radiomics", "texture", and "delta". Studies were analysed using QUADAS-2 and the RQS tool.

RESULTS

Forty-eight studies were finally included. The studies were divided into preclinical/methodological (five studies, 10.4%); rectal cancer (six studies, 12.5%); lung cancer (twelve studies, 25%); sarcoma (five studies, 10.4%); prostate cancer (three studies, 6.3%), head and neck cancer (six studies, 12.5%); gastrointestinal malignancies excluding rectum (seven studies, 14.6%), and other disease sites (four studies, 8.3%). The median RQS of all studies was 25% (mean 21% ± 12%), with 13 studies (30.2%) achieving a quality score < 10% and 22 studies (51.2%) < 25%.

CONCLUSIONS

Delta radiomics shows potential benefit for several clinical endpoints in oncology (differential diagnosis, prognosis and prediction of treatment response, and evaluation of side effects). Nevertheless, the studies included in this systematic review suffer from the bias of overall low quality, so that the conclusions are currently heterogeneous, not robust, and not replicable. Further research with prospective and multicentre studies is needed for the clinical validation of delta radiomics approaches.

Transfer learning for fluence map prediction in adrenal stereotactic body radiation therapy

Objective:To design a deep transfer learning framework for modeling fluence map predictions for stereotactic body radiation therapy (SBRT) of adrenal cancer and similar sites that usually have a small number of cases.Approach:We developed a transfer learning framework for adrenal SBRT planning that leverages knowledge in a pancreas SBRT planning model. Treatment plans from the two sites had different dose prescriptions and beam settings but both prioritized gastrointestinal sparing. A base framework was first trained with 100 pancreas cases. This framework consists of two convolutional neural networks (CNN), which predict individual beam doses (BD-CNN) and fluence maps (FM-CNN) sequentially for 9-beam intensity-modulated radiation therapy (IMRT) plans. Forty-five adrenal plans were split into training/validation/test sets with the ratio of 20/10/15. The base BD-CNN was re-trained with transfer learning using 5/10/15/20 adrenal training cases to produce multiple candidate adrenal BD-CNN models. The base FM-CNN was directly used for adrenal cases. The deep learning (DL) plans were evaluated by several clinically relevant dosimetric endpoints, producing a percentage score relative to the clinical plans.Main results:Transfer learning significantly reduced the number of training cases and training time needed to train such a DL framework. The adrenal transfer learning model trained with 5/10/15/20 cases achieved validation plan scores of 85.4/91.2/90.7/89.4, suggesting that model performance saturated with 10 training cases. Meanwhile, a model using all 20 adrenal training cases without transfer learning only scored 80.5. For the final test set, the 5/10/15/20-case models achieved scores of 73.5/75.3/78.9/83.3.Significance:It is feasible to use deep transfer learning to train an IMRT fluence prediction framework. This technique could adapt to different dose prescriptions and beam configurations. This framework potentially enables DL modeling for clinical sites that have a limited dataset, either due to few cases or due to rapid technology evolution.

Pulmonary lesion subtypes recognition of COVID-19 from radiomics data with three-dimensional texture characterization in computed tomography images

BACKGROUND

The COVID-19 disease is putting unprecedented pressure on the global healthcare system. The CT (computed tomography) examination as a auxiliary confirmed diagnostic method can help clinicians quickly detect lesions locations of COVID-19 once screening by PCR test. Furthermore, the lesion subtypes classification plays a critical role in the consequent treatment decision. Identifying the subtypes of lesions accurately can help doctors discover changes in lesions in time and better assess the severity of COVID-19.

METHOD

The most four typical lesion subtypes of COVID-19 are discussed in this paper, which are GGO (ground-glass opacity), cord, solid and subsolid. A computer-aided diagnosis approach of lesion subtype is proposed in this paper. The radiomics data of lesions are segmented from COVID-19 patients CT images with diagnosis and lesions annotations by radiologists. Then the three-dimensional texture descriptors are applied on the volume data of lesions as well as shape and first-order features. The massive feature data are selected by HAFS (hybrid adaptive feature selection) algorithm and a classification model is trained at the same time. The classifier is used to predict lesion subtypes as side decision information for radiologists.

RESULTS

There are 3734 lesions extracted from the dataset with 319 patients collection and then 189 radiomics features are obtained finally. The random forest classifier is trained with data augmentation that the number of different subtypes of lesions is imbalanced in initial dataset. The experimental results show that the accuracy of the four subtypes of lesions is (93.06%, 96.84%, 99.58%, and 94.30%), the recall is (95.52%, 91.58%, 95.80% and 80.75%) and the f-score is (93.84%, 92.37%, 95.47%, and 84.42%).

CONCLUSION

The three-dimensional radiomics features used in this paper can better express the high-level information of COVID-19 lesions in CT slices. HAFS method aggregates the results of multiple feature selection algorithms intersects with traditional methods to filter out redundant features more accurately. After selection, the subtype of COVID-19 lesion can be judged by inputting the features into the RF (random forest) model, which can help clinicians more accurately identify the subtypes of COVID-19 lesions and provide help for further research.

Evaluation of prediction and classification performances in different machine learning models for patient-specific quality assurance of head-and-neck VMAT plans

PURPOSE

The purpose of this study is to evaluate the prediction and classification performances of the gamma passing rate (GPR) for different machine learning models and to select the best model for achieving machine learning-based patient-specific quality assurance (PSQA).

METHODS

The measurement verification of 356 head-and-neck volumetric modulated arc therapy plans was performed using a diode array phantom (Delta4 Phantom), and GPR values at 2%/2 mm with global normalization and 3%/2 mm with local normalization were calculated. Machine learning models, including ridge regression (RIDGE), random forest (RF), support vector regression (SVR), and stacked generalization (STACKING), were used to predict the GPR. Each machine learning model was trained using 260 plans, and the prediction accuracy was evaluated using the remaining 96 plans. The prediction error between the measured and predicted GPR was evaluated. For the classification evaluation, the lower control limit for the measured GPR and lower control limit for predicted GPR (LCL_p ) was defined to identify whether the GPR values represent a "pass" or a "fail." LCL_p values with 99% and 99.9% confidence levels were calculated as the upper prediction limits for the GPR estimated from the linear regression between the measured and predicted GPR.

RESULTS

There was an overestimation trend of the low measured GPR. The maximum prediction errors for RIDGE, RF, SVR, and STACKING were 3.2%, 2.9%, 2.3%, and 2.2% at the global 2%/2 mm and 6.3%, 6.6%, 6.1%, and 5.5% at the local 3%/2 mm, respectively. In the global 2%/2 mm, the sensitivity was 100% for all the machine learning models except RIDGE when using 99% LCL_p . The specificity was 76.1% for RIDGE, RF, and SVR and 66.3% for STACKING; however, the specificity decreased dramatically when 99.9% LCL_p was used. In the local 3%/2 mm, however, only STACKING showed 100% sensitivity when using 99% LCL_p . The decrease in the specificity using 99.9% LCL_p was smaller than that in the global 2%/2 mm, and the specificity for RIDGE, RF, SVR, and STACKING was 61.3%, 61.3%, 72.0%, and 66.8%, respectively.

CONCLUSIONS

STACKING had better prediction accuracy for low GPR values than other machine learning models. Applying LCL_p to a regression model enabled the consistent evaluation of quantitative and qualitative GPR predictions. Adjusting the confidence level of the LCL_p helped improve the balance between the sensitivity and specificity. We suggest that STACKING can assist the safe and efficient operation of PSQA.

A Comprehensive Survey of Image-Based Food Recognition and Volume Estimation Methods for Dietary Assessment

Dietary studies showed that dietary problems such as obesity are associated with other chronic diseases, including hypertension, irregular blood sugar levels, and increased risk of heart attacks. The primary cause of these problems is poor lifestyle choices and unhealthy dietary habits, which are manageable using interactive mHealth apps. However, traditional dietary monitoring systems using manual food logging suffer from imprecision, underreporting, time consumption, and low adherence. Recent dietary monitoring systems tackle these challenges by automatic assessment of dietary intake through machine learning methods. This survey discusses the best-performing methodologies that have been developed so far for automatic food recognition and volume estimation. Firstly, the paper presented the rationale of visual-based methods for food recognition. Then, the core of the study is the presentation, discussion, and evaluation of these methods based on popular food image databases. In this context, this study discusses the mobile applications that are implementing these methods for automatic food logging. Our findings indicate that around 66.7% of surveyed studies use visual features from deep neural networks for food recognition. Similarly, all surveyed studies employed a variant of convolutional neural networks (CNN) for ingredient recognition due to recent research interest. Finally, this survey ends with a discussion of potential applications of food image analysis, existing research gaps, and open issues of this research area. Learning from unlabeled image datasets in an unsupervised manner, catastrophic forgetting during continual learning, and improving model transparency using explainable AI are potential areas of interest for future studies.

Utility of a Machine-Guided Tool for Assessing Risk Behavior Associated With Contracting HIV in Three Sites in South Africa: Protocol for an In-Field Evaluation

BACKGROUND

Mobile technology has helped to advance health programs, and studies have shown that an automated risk prediction model can successfully be used to identify patients who exhibit a high probable risk of contracting human immunodeficiency virus (HIV). A machine-guided tool is an algorithm that takes a set of subjective and objective answers from a simple questionnaire and computes an HIV risk assessment score.

OBJECTIVE

The primary objective of this study is to establish that machine learning can be used to develop machine-guided tools and give us a deeper statistical understanding of the correlation between certain behavioral patterns and HIV.

METHODS

In total, 200 HIV-negative adult individuals across three South African study sites each (two semirural and one urban) will be recruited. Study processes will include (1) completing a series of questions (demographic, sexual behavior and history, personal, lifestyle, and symptoms) on an application system, unaided (assistance will only be provided upon user request); (2) two HIV tests (one per study visit) being performed by a nurse/counselor according to South African national guidelines (to evaluate the prediction accuracy of the tool); and (3) communicating test results and completing a user experience survey questionnaire. The output metrics for this study will be computed by using the participants' risk assessment scores as "predictions" and the test results as the "ground truth." Analyses will be completed after visit 1 and then again after visit 2. All risk assessment scores will be used to calculate the reliability of the machine-guided tool.

RESULTS

Ethical approval was received from the University of Witwatersrand Human Research Ethics Committee (HREC; ethics reference no. 200312) on August 20, 2020. This study is ongoing. Data collection has commenced and is expected to be completed in the second half of 2021. We will report on the machine-guided tool's performance and usability, together with user satisfaction and recommendations for improvement.

CONCLUSIONS

Machine-guided risk assessment tools can provide a cost-effective alternative to large-scale HIV screening and help in providing targeted counseling and testing to prevent the spread of HIV.

TRIAL REGISTRATION

South African National Clinical Trial Registry DOH-27-042021-679; https://sanctr.samrc.ac.za/TrialDisplay.aspx?TrialID=5545.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/30304.

How pronunciation distance impacts word recognition in children and adults

Although unfamiliar accents can pose word identification challenges for children and adults, few studies have directly compared perception of multiple nonnative and regional accents or quantified how the extent of deviation from the ambient accent impacts word identification accuracy across development. To address these gaps, 5- to 7-year-old children's and adults' word identification accuracy with native (Midland American, British, Scottish), nonnative (German-, Mandarin-, Japanese-accented English) and bilingual (Hindi-English) varieties (one talker per accent) was tested in quiet and noise. Talkers' pronunciation distance from the ambient dialect was quantified at the phoneme level using a Levenshtein algorithm adaptation. Whereas performance was worse on all non-ambient dialects than the ambient one, there were only interactions between talker and age (child vs adult or across age for the children) for a subset of talkers, which did not fall along the native/nonnative divide. Levenshtein distances significantly predicted word recognition accuracy for adults and children in both listening environments with similar impacts in quiet. In noise, children had more difficulty overcoming pronunciations that substantially deviated from ambient dialect norms than adults. Future work should continue investigating how pronunciation distance impacts word recognition accuracy by incorporating distance metrics at other levels of analysis (e.g., phonetic, suprasegmental).

A deep learning approach using effective preprocessing techniques to detect COVID-19 from chest CT-scan and X-ray images

Coronavirus disease-19 (COVID-19) is a severe respiratory viral disease first reported in late 2019 that has spread worldwide. Although some wealthy countries have made significant progress in detecting and containing this disease, most underdeveloped countries are still struggling to identify COVID-19 cases in large populations. With the rising number of COVID-19 cases, there are often insufficient COVID-19 diagnostic kits and related resources in such countries. However, other basic diagnostic resources often do exist, which motivated us to develop Deep Learning models to assist clinicians and radiologists to provide prompt diagnostic support to the patients. In this study, we have developed a deep learning-based COVID-19 case detection model trained with a dataset consisting of chest CT scans and X-ray images. A modified ResNet50V2 architecture was employed as deep learning architecture in the proposed model. The dataset utilized to train the model was collected from various publicly available sources and included four class labels: confirmed COVID-19, normal controls and confirmed viral and bacterial pneumonia cases. The aggregated dataset was preprocessed through a sharpening filter before feeding the dataset into the proposed model. This model attained an accuracy of 96.452% for four-class cases (COVID-19/Normal/Bacterial pneumonia/Viral pneumonia), 97.242% for three-class cases (COVID-19/Normal/Bacterial pneumonia) and 98.954% for two-class cases (COVID-19/Viral pneumonia) using chest X-ray images. The model acquired a comprehensive accuracy of 99.012% for three-class cases (COVID-19/Normal/Community-acquired pneumonia) and 99.99% for two-class cases (Normal/COVID-19) using CT-scan images of the chest. This high accuracy presents a new and potentially important resource to enable radiologists to identify and rapidly diagnose COVID-19 cases with only basic but widely available equipment.

Artificial intelligence in image-guided radiotherapy: a review of treatment target localization

Modern conformal beam delivery techniques require image-guidance to ensure the prescribed dose to be delivered as planned. Recent advances in artificial intelligence (AI) have greatly augmented our ability to accurately localize the treatment target while sparing the normal tissues. In this paper, we review the applications of AI-based algorithms in image-guided radiotherapy (IGRT), and discuss the indications of these applications to the future of clinical practice of radiotherapy. The benefits, limitations and some important trends in research and development of the AI-based IGRT techniques are also discussed. AI-based IGRT techniques have the potential to monitor tumor motion, reduce treatment uncertainty and improve treatment precision. Particularly, these techniques also allow more healthy tissue to be spared while keeping tumor coverage the same or even better.

pySuStaIn: a Python implementation of the Subtype and Stage Inference algorithm

Progressive disorders are highly heterogeneous. Symptom-based clinical classification of these disorders may not reflect the underlying pathobiology. Data-driven subtyping and staging of patients has the potential to disentangle the complex spatiotemporal patterns of disease progression. Tools that enable this are in high demand from clinical and treatment-development communities. Here we describe the pySuStaIn software package, a Python-based implementation of the Subtype and Stage Inference (SuStaIn) algorithm. SuStaIn unravels the complexity of heterogeneous diseases by inferring multiple disease progression patterns (subtypes) and individual severity (stages) from cross-sectional data. The primary aims of pySuStaIn are to enable widespread application and translation of SuStaIn via an accessible Python package that supports simple extension and generalization to novel modelling situations within a single, consistent architecture.

Overview of Explainable Artificial Intelligence for Prognostic and Health Management of Industrial Assets Based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses

Surveys on explainable artificial intelligence (XAI) are related to biology, clinical trials, fintech management, medicine, neurorobotics, and psychology, among others. Prognostics and health management (PHM) is the discipline that links the studies of failure mechanisms to system lifecycle management. There is a need, which is still absent, to produce an analytical compilation of PHM-XAI works. In this paper, we use preferred reporting items for systematic reviews and meta-analyses (PRISMA) to present a state of the art on XAI applied to PHM of industrial assets. This work provides an overview of the trend of XAI in PHM and answers the question of accuracy versus explainability, considering the extent of human involvement, explanation assessment, and uncertainty quantification in this topic. Research articles associated with the subject, since 2015 to 2021, were selected from five databases following the PRISMA methodology, several of them related to sensors. The data were extracted from selected articles and examined obtaining diverse findings that were synthesized as follows. First, while the discipline is still young, the analysis indicates a growing acceptance of XAI in PHM. Second, XAI offers dual advantages, where it is assimilated as a tool to execute PHM tasks and explain diagnostic and anomaly detection activities, implying a real need for XAI in PHM. Third, the review shows that PHM-XAI papers provide interesting results, suggesting that the PHM performance is unaffected by the XAI. Fourth, human role, evaluation metrics, and uncertainty management are areas requiring further attention by the PHM community. Adequate assessment metrics to cater to PHM needs are requested. Finally, most case studies featured in the considered articles are based on real industrial data, and some of them are related to sensors, showing that the available PHM-XAI blends solve real-world challenges, increasing the confidence in the artificial intelligence models' adoption in the industry.

Therapeutic cancer vaccines revamping: technology advancements and pitfalls

Cancer vaccines (CVs) represent a long-sought therapeutic and prophylactic immunotherapy strategy to obtain antigen (Ag)-specific T-cell responses and potentially achieve long-term clinical benefit. However, historically, most CV clinical trials have resulted in disappointing outcomes, despite promising signs of immunogenicity across most formulations. In the past decade, technological advances regarding vaccine delivery platforms, tools for immunogenomic profiling, and Ag/epitope selection have occurred. Consequently, the ability of CVs to induce tumor-specific and, in some cases, remarkable clinical responses have been observed in early-phase clinical trials. It is notable that the record-breaking speed of vaccine development in response to the coronavirus disease-2019 pandemic mainly relied on manufacturing infrastructures and technological platforms already developed for CVs. In turn, research, clinical data, and infrastructures put in place for the severe acute respiratory syndrome coronavirus 2 pandemic can further speed CV development processes. This review outlines the main technological advancements as well as major issues to tackle in the development of CVs. Possible applications for unmet clinical needs will be described, putting into perspective the future of cancer vaccinology.

Artificial intelligence applications in intensity modulated radiation treatment planning: an overview

Artificial intelligence (AI) refers to methods that improve and automate challenging human tasks by systematically capturing and applying relevant knowledge in these tasks. Over the past decades, a number of approaches have been developed to address different types and needs of system intelligence ranging from search strategies to knowledge representation and inference to robotic planning. In the context of radiation treatment planning, multiple AI approaches may be adopted to improve the planning quality and efficiency. For example, knowledge representation and inference methods may improve dose prescription by integrating and reasoning about the domain knowledge described in many clinical guidelines and clinical trials reports. In this review, we will focus on the most studied AI approach in intensity modulated radiation therapy (IMRT)/volumetric modulated arc therapy (VMAT)-machine learning (ML) and describe our recent efforts in applying ML to improve the quality, consistency, and efficiency of IMRT/VMAT planning. With the available high-quality data, we can build models to accurately predict critical variables for each step of the planning process and thus automate and improve its outcomes. Specific to the IMRT/VMAT planning process, we can build models for each of the four critical components in the process: dose-volume histogram (DVH), Dose, Fluence, and Human Planner. These models can be divided into two general groups. The first group focuses on encoding prior experience and knowledge through ML and more recently deep learning (DL) from prior clinical plans and using these models to predict the optimal DVH (DVH prediction model), or 3D dose distribution (dose prediction model), or fluence map (fluence map model). The goal of these models is to reduce or remove the trial-and-error process and guarantee consistently high-quality plans. The second group of models focuses on mimicking human planners' decision-making process (planning strategy model) during the iterative adjustments/guidance of the optimization engine. Each critical step of the IMRT/VMAT treatment planning process can be improved and automated by AI methods. As more training data becomes available and more sophisticated models are developed, we can expect that the AI methods in treatment planning will continue to improve accuracy, efficiency, and robustness.

Advances in the computational landscape for repurposed drugs against COVID-19

The COVID-19 pandemic has caused millions of deaths and massive societal distress worldwide. Therapeutic solutions are urgently needed, but de novo drug development remains a lengthy process. One promising alternative is computational drug repurposing, which enables the prioritization of existing compounds through fast in silico analyses. Recent efforts based on molecular docking, machine learning, and network analysis have produced actionable predictions. Some predicted drugs, targeting viral proteins and pathological host pathways are undergoing clinical trials. Here, we review this work, highlight drugs with high predicted efficacy and classify their mechanisms of action. We discuss the strengths and limitations of the published methodologies and outline possible future directions. Finally, we curate a list of COVID-19 data portals and other repositories that could be used to accelerate future research.

A preliminary analysis of AI based smartphone application for diagnosis of COVID-19 using chest X-ray images

The COVID-19 outbreak has catastrophically affected both public health system and world economy. Swift diagnosis of the positive cases will help in providing proper medical attention to the infected individuals and will also aid in effective tracing of their contacts to break the chain of transmission. Blending Artificial Intelligence (AI) with chest X-ray images and incorporating these models in a smartphone can be handy for the accelerated diagnosis of COVID-19. In this study, publicly available datasets of chest X-ray images have been utilized for training and testing of five pre-trained Convolutional Neural Network (CNN) models namely VGG16, MobileNetV2, Xception, NASNetMobile and InceptionResNetV2. Prior to the training of the selected models, the number of images in COVID-19 category has been increased employing traditional augmentation and Generative Adversarial Network (GAN). The performance of the five pre-trained CNN models utilizing the images generated with the two strategies has been compared. In the case of models trained using augmented images, Xception (98%) and MobileNetV2 (97.9%) turned out to be the ones with highest validation accuracy. Xception (98.1%) and VGG16 (98.6%) emerged as models with the highest validation accuracy in the models trained with synthetic GAN images. The best performing models have been further deployed in a smartphone and evaluated. The overall results suggest that VGG16 and Xception, trained with the synthetic images created using GAN, performed better compared to models trained with augmented images. Among these two models VGG16 produced an encouraging Diagnostic Odd Ratio (DOR) with higher positive likelihood and lower negative likelihood for the prediction of COVID-19.

Deep Learning Algorithms Achieved Satisfactory Predictions When Trained on a Novel Collection of Anticoronavirus Molecules

Drug discovery and repurposing against COVID-19 is a highly relevant topic with huge efforts dedicated to delivering novel therapeutics targeting SARS-CoV-2. In this context, computer-aided drug discovery is of interest in orienting the early high throughput screenings and in optimizing the hit identification rate. We herein propose a pipeline for Ligand-Based Drug Discovery (LBDD) against SARS-CoV-2. Through an extensive search of the literature and multiple steps of filtering, we integrated information on 2,610 molecules having a validated effect against SARS-CoV and/or SARS-CoV-2. The chemical structures of these molecules were encoded through multiple systems to be readily useful as input to conventional machine learning (ML) algorithms or deep learning (DL) architectures. We assessed the performances of seven ML algorithms and four DL algorithms in achieving molecule classification into two classes: active and inactive. The Random Forests (RF), Graph Convolutional Network (GCN), and Directed Acyclic Graph (DAG) models achieved the best performances. These models were further optimized through hyperparameter tuning and achieved ROC-AUC scores through cross-validation of 85, 83, and 79% for RF, GCN, and DAG models, respectively. An external validation step on the FDA-approved drugs collection revealed a superior potential of DL algorithms to achieve drug repurposing against SARS-CoV-2 based on the dataset herein presented. Namely, GCN and DAG achieved more than 50% of the true positive rate assessed on the confirmed hits of a PubChem bioassay.

Real-World Evidence of COVID-19 Patients' Data Quality in the Electronic Health Records

Despite the importance of electronic health records data, less attention has been given to data quality. This study aimed to evaluate the quality of COVID-19 patients' records and their readiness for secondary use. We conducted a retrospective chart review study of all COVID-19 inpatients in an academic healthcare hospital for the year 2020, which were identified using ICD-10 codes and case definition guidelines. COVID-19 signs and symptoms were higher in unstructured clinical notes than in structured coded data. COVID-19 cases were categorized as 218 (66.46%) "confirmed cases", 10 (3.05%) "probable cases", 9 (2.74%) "suspected cases", and 91 (27.74%) "no sufficient evidence". The identification of "probable cases" and "suspected cases" was more challenging than "confirmed cases" where laboratory confirmation was sufficient. The accuracy of the COVID-19 case identification was higher in laboratory tests than in ICD-10 codes. When validating using laboratory results, we found that ICD-10 codes were inaccurately assigned to 238 (72.56%) patients' records. "No sufficient evidence" records might indicate inaccurate and incomplete EHR data. Data quality evaluation should be incorporated to ensure patient safety and data readiness for secondary use research and predictive analytics. We encourage educational and training efforts to motivate healthcare providers regarding the importance of accurate documentation at the point-of-care.

Reflections on epistemological aspects of artificial intelligence during the COVID-19 pandemic

Artificial intelligence plays an important role and has been used by several countries as a health strategy in an attempt to understand, control and find a cure for the disease caused by Coronavirus. These intelligent systems can assist in accelerating the process of developing antivirals for Coronavirus and in predicting new variants of this virus. For this reason, much research on COVID-19 has been developed with the aim of contributing to new discoveries about the Coronavirus. However, there are some epistemological aspects about the use of AI in this pandemic period of Covid-19 that deserve to be discussed and need reflections. In this scenario, this article presents a reflection on the two epistemological aspects faced by the COVID-19 pandemic: (1) The epistemological aspect resulting from the use of patient data to fill the knowledge base of intelligent systems; (2) the epistemological problem arising from the dependence of health professionals on the results/diagnoses issued by intelligent systems. In addition, we present some epistemological challenges to be implemented in a pandemic period.

What Drives Bitcoin? An Approach from Continuous Local Transfer Entropy and Deep Learning Classification Models

Bitcoin has attracted attention from different market participants due to unpredictable price patterns. Sometimes, the price has exhibited big jumps. Bitcoin prices have also had extreme, unexpected crashes. We test the predictive power of a wide range of determinants on bitcoins’ price direction under the continuous transfer entropy approach as a feature selection criterion. Accordingly, the statistically significant assets in the sense of permutation test on the nearest neighbour estimation of local transfer entropy are used as features or explanatory variables in a deep learning classification model to predict the price direction of bitcoin. The proposed variable selection do not find significative the explanatory power of NASDAQ and Tesla. Under different scenarios and metrics, the best results are obtained using the significant drivers during the pandemic as validation. In the test, the accuracy increased in the post-pandemic scenario of July 2020 to January 2021 without drivers. In other words, our results indicate that in times of high volatility, Bitcoin seems to self-regulate and does not need additional drivers to improve the accuracy of the price direction.

Insights of an AI agent via analysis of prediction errors: a case study of fluence map prediction for radiation therapy planning

Purpose.We have previously reported an artificial intelligence (AI) agent that automatically generates intensity-modulated radiation therapy (IMRT) plans via fluence map prediction, by-passing inverse planning. This AI agent achieved clinically comparable quality for prostate cases, but its performance on head-and-neck patients leaves room for improvement. This study aims to collect insights of the deep-learning-based (DL-based) fluence map prediction model by systematically analyzing its prediction errors.Methods.From the modeling perspective, the DL model's output is the fluence maps of IMRT plans. However, from the clinical planning perspective, the plan quality evaluation should be based on the clinical dosimetric criteria such as dose-volume histograms. To account for the complex and non-intuitive relationships between fluence map prediction errors and the corresponding dose distribution changes, we propose a novel error analysis approach that systematically examines plan dosimetric changes that are induced by varying amounts of fluence prediction errors. We investigated four decomposition modes of model prediction errors. The two spatial domain decompositions are based on fluence intensity and fluence gradient. The two frequency domain decompositions are based on Fourier-space banded frequency rings and Fourier-space truncated low-frequency disks. The decomposed error was analyzed for its impact on the resulting plans' dosimetric metrics. The analysis was conducted on 15 test cases spared from the 200 training and 16 validation cases used to train the model.Results.Most planning target volume metrics were significantly correlated with most error decompositions. The Fourier space disk radii had the largest Spearman's coefficients. The low-frequency region within a disk of ∼20% Fourier space contained most of errors that impact overall plan quality.Conclusions.This study demonstrates the feasibility of using fluence map prediction error analysis to understand the AI agent's performance. Such insights will help fine-tune the DL models in architecture design and loss function selection.

Data-Driven Discovery of Mathematical and Physical Relations in Oncology Data Using Human-Understandable Machine Learning

For decades, researchers have used the concepts of rate of change and differential equations to model and forecast neoplastic processes. This expressive mathematical apparatus brought significant insights in oncology by describing the unregulated proliferation and host interactions of cancer cells, as well as their response to treatments. Now, these theories have been given a new life and found new applications. With the advent of routine cancer genome sequencing and the resulting abundance of data, oncology now builds an "arsenal" of new modeling and analysis tools. Models describing the governing physical laws of tumor-host-drug interactions can be now challenged with biological data to make predictions about cancer progression. Our study joins the efforts of the mathematical and computational oncology community by introducing a novel machine learning system for data-driven discovery of mathematical and physical relations in oncology. The system utilizes computational mechanisms such as competition, cooperation, and adaptation in neural networks to simultaneously learn the statistics and the governing relations between multiple clinical data covariates. Targeting an easy adoption in clinical oncology, the solutions of our system reveal human-understandable properties and features hidden in the data. As our experiments demonstrate, our system can describe nonlinear conservation laws in cancer kinetics and growth curves, symmetries in tumor's phenotypic staging transitions, the preoperative spatial tumor distribution, and up to the nonlinear intracellular and extracellular pharmacokinetics of neoadjuvant therapies. The primary goal of our work is to enhance or improve the mechanistic understanding of cancer dynamics by exploiting heterogeneous clinical data. We demonstrate through multiple instantiations that our system is extracting an accurate human-understandable representation of the underlying dynamics of physical interactions central to typical oncology problems. Our results and evaluation demonstrate that, using simple-yet powerful-computational mechanisms, such a machine learning system can support clinical decision-making. To this end, our system is a representative tool of the field of mathematical and computational oncology and offers a bridge between the data, the modeler, the data scientist, and the practicing clinician.

Trustworthy Artificial Intelligence in Medical Imaging

Trust in artificial intelligence (AI) by society and the development of trustworthy AI systems and ecosystems are critical for the progress and implementation of AI technology in medicine. With the growing use of AI in a variety of medical and imaging applications, it is more vital than ever to make these systems dependable and trustworthy. Fourteen core principles are considered in this article aiming to move the needle more closely to systems that are accurate, resilient, fair, explainable, safe, and transparent: toward trustworthy AI.

Effective transfer entropy to measure information flows in credit markets

In this paper we propose to study the dynamics of financial contagion between the credit default swap (CDS) and the sovereign bond markets through effective transfer entropy, a model-free methodology which enables to overcome the required hypotheses of classical price discovery measures in the statistical and econometric literature, without being restricted to linear dynamics. By means of effective transfer entropy we correct for small sample biases which affect the traditional Shannon transfer entropy, as well as we are able to conduct inference on the estimated directional information flows. In our empirical application, we analyze the CDS and bond market data for eight countries of the European Union, and aim to discover which of the two assets is faster at incorporating the information on the credit risk of the underlying sovereign. Our results show a clear and statistically significant prominence of the bond market for pricing the sovereign credit risk, especially during the crisis period. During the post-crisis period, instead, a few countries behave dissimilarly from the others, in particular Spain and the Netherlands.

Early Stage Identification of COVID-19 Patients in Mexico Using Machine Learning: A Case Study for the Tijuana General Hospital

Background: The current pandemic caused by SARS-CoV-2 is an acute illness of global concern. SARS-CoV-2 is an infectious disease caused by a recently discovered coronavirus. Most people who get sick from COVID-19 experience either mild, moderate, or severe symptoms. In order to help make quick decisions regarding treatment and isolation needs, it is useful to determine which significant variables indicate infection cases in the population served by the Tijuana General Hospital (Hospital General de Tijuana). An Artificial Intelligence (Machine Learning) mathematical model was developed in order to identify early-stage significant variables in COVID-19 patients. Methods: The individual characteristics of the study subjects included age, gender, age group, symptoms, comorbidities, diagnosis, and outcomes. A mathematical model that uses supervised learning algorithms, allowing the identification of the significant variables that predict the diagnosis of COVID-19 with high precision, was developed. Results: Automatic algorithms were used to analyze the data: for Systolic Arterial Hypertension (SAH), the Logistic Regression algorithm showed results of 91.0% in area under ROC (AUC), 80% accuracy (CA), 80% F1 and 80% Recall, and 80.1% precision for the selected variables, while for Diabetes Mellitus (DM) with the Logistic Regression algorithm it obtained 91.2% AUC, 89.2% accuracy, 88.8% F1, 89.7% precision, and 89.2% recall for the selected variables. The neural network algorithm showed better results for patients with Obesity, obtaining 83.4% AUC, 91.4% accuracy, 89.9% F1, 90.6% precision, and 91.4% recall. Conclusions: Statistical analyses revealed that the significant predictive symptoms in patients with SAH, DM, and Obesity were more substantial in fatigue and myalgias/arthralgias. In contrast, the third dominant symptom in people with SAH and DM was odynophagia.

Artificial Intelligence in Medical Imaging and its Impact on the Rare Disease Community: Threats, Challenges and Opportunities

Almost 1 in 10 individuals can suffer from one of many rare diseases (RDs). The average time to diagnosis for an RD patient is as high as 7 years. Artificial intelligence (AI)-based positron emission tomography (PET), if implemented appropriately, has tremendous potential to advance the diagnosis of RDs. Patient advocacy groups must be active stakeholders in the AI ecosystem if we are to avoid potential issues related to the implementation of AI into health care. AI medical devices must not only be RD-aware at each stage of their conceptualization and life cycle but also should be trained on diverse and augmented datasets representative of the end-user population including RDs. Inability to do so leads to potential harm and unsustainable deployment of AI-based medical devices (AIMDs) into clinical practice.

Topological Data Analysis as a New Tool for EEG Processing

Electroencephalography (EEG) is a widely used cerebral activity measuring device for both clinical and everyday life applications. In addition to denoising and potential classification, a crucial step in EEG processing is to extract relevant features. Topological data analysis (TDA) as an emerging tool enables to analyse and understand data from a different angle than traditionally used methods. As a higher dimensional analogy of graph analysis, TDA can model rich interactions beyond pairwise relations. It also distinguishes different dynamics of EEG time series. TDA remains largely unknown to the EEG processing community while it fits well the heterogeneous nature of EEG signals. This short review aims to give a quick introduction to TDA and how it can be applied to EEG analysis in various applications including brain-computer interfaces (BCIs). After introducing the objective of the article, the main concepts and ideas of TDA are explained. Next, how to implement it for EEG processing is detailed, and lastly the article discusses the benefits and limitations of the method.

Translating the Machine: Skills that Human Clinicians Must Develop in the Era of Artificial Intelligence

In coming decades, artificial intelligence (AI) platforms are expected to build on the profound achievements demonstrated in research papers towards implementation in real-world medicine. The implementation of AI systems is likely to be as an adjunct to clinicians rather than a replacement, but it still has the potential for a revolutionary impact on ophthalmology specifically and medicine in general in terms of addressing crucial scientific, socioeconomic and capacity challenges facing populations worldwide. In this paper we discuss the broad range of skills that clinicians should develop or refine to be able to fully embrace the opportunities that this technology will bring. We highlight the need for an awareness to identify AI systems that might already be in place and the need to be able to properly assess the utility of their outputs to correctly incorporate the AI system into clinical workflows. In a second section we discuss the need for clinicians to cultivate those human skills that are beyond the capabilities of the AI platforms and which should be just as important as ever. We describe the need for such an awareness by providing clinical examples of situations that might in the future arise in human interactions with machine algorithms. We also envisage a harmonious future in which an educated human and machine interaction can be optimised for the best possible patient experience and care.

Enhancing the impact of Artificial Intelligence in Medicine: A joint AIFM-INFN Italian initiative for a dedicated cloud-based computing infrastructure

Artificial Intelligence (AI) techniques have been implemented in the field of Medical Imaging for more than forty years. Medical Physicists, Clinicians and Computer Scientists have been collaborating since the beginning to realize software solutions to enhance the informative content of medical images, including AI-based support systems for image interpretation. Despite the recent massive progress in this field due to the current emphasis on Radiomics, Machine Learning and Deep Learning, there are still some barriers to overcome before these tools are fully integrated into the clinical workflows to finally enable a precision medicine approach to patients' care. Nowadays, as Medical Imaging has entered the Big Data era, innovative solutions to efficiently deal with huge amounts of data and to exploit large and distributed computing resources are urgently needed. In the framework of a collaboration agreement between the Italian Association of Medical Physicists (AIFM) and the National Institute for Nuclear Physics (INFN), we propose a model of an intensive computing infrastructure, especially suited for training AI models, equipped with secure storage systems, compliant with data protection regulation, which will accelerate the development and extensive validation of AI-based solutions in the Medical Imaging field of research. This solution can be developed and made operational by Physicists and Computer Scientists working on complementary fields of research in Physics, such as High Energy Physics and Medical Physics, who have all the necessary skills to tailor the AI-technology to the needs of the Medical Imaging community and to shorten the pathway towards the clinical applicability of AI-based decision support systems.

Back to square one: the bodily roots of conscious experiences in early life

Most theoretical and empirical discussions about the nature of consciousness are typically couched in a way that endorses a tacit adult-centric and vision-based perspective. This paper defends the idea that consciousness science may be put on a fruitful track for its next phase by examining the nature of subjective experiences through a bottom-up developmental lens. We draw attention to the intrinsic link between consciousness, experiences and experiencing subjects, which are first and foremost embodied and situated organisms essentially concerned with self-preservation within a precarious environment. Our paper suggests that in order to understand what consciousness 'is', one should first tackle the fundamental question: how do embodied experiences 'arise' from square one? We then highlight one key yet overlooked aspect of human consciousness studies, namely that the earliest and closest environment of an embodied experiencing subject is the body of another human experiencing subject. We present evidence speaking in favour of fairly sophisticated forms of early sensorimotor integration of bodily signals and self-generated actions already being established in utero. We conclude that these primitive and fundamentally relational and co-embodied roots of our early experiences may have a crucial impact on the way human beings consciously experience the self, body and the world across their lifespan.

Social Networks of Lexical Innovation. Investigating the Social Dynamics of Diffusion of Neologisms on Twitter

Societies continually evolve and speakers use new words to talk about innovative products and practices. While most lexical innovations soon fall into disuse, others spread successfully and become part of the lexicon. In this paper, I conduct a longitudinal study of the spread of 99 English neologisms on Twitter to study their degrees and pathways of diffusion. Previous work on lexical innovation has almost exclusively relied on usage frequency for investigating the spread of new words. To get a more differentiated picture of diffusion, I use frequency-based measures to study temporal aspects of diffusion and I use network analyses for a more detailed and accurate investigation of the sociolinguistic dynamics of diffusion. The results show that frequency measures manage to capture diffusion with varying success. Frequency counts can serve as an approximate indicator for overall degrees of diffusion, yet they miss important information about the temporal usage profiles of lexical innovations. The results indicate that neologisms with similar total frequency can exhibit significantly different degrees of diffusion. Analysing differences in their temporal dynamics of use with regard to their age, trends in usage intensity, and volatility contributes to a more accurate account of their diffusion. The results obtained from the social network analysis reveal substantial differences in the social pathways of diffusion. Social diffusion significantly correlates with the frequency and temporal usage profiles of neologisms. However, the network visualisations and metrics identify neologisms whose degrees of social diffusion are more limited than suggested by their overall frequency of use. These include, among others, highly volatile neologisms (e.g., poppygate) and political terms (e.g., alt-left), whose use almost exclusively goes back to single communities of closely-connected, like-minded individuals. I argue that the inclusion of temporal and social information is of particular importance for the study of lexical innovation since neologisms exhibit high degrees of temporal volatility and social indexicality. More generally, the present approach demonstrates the potential of social network analysis for sociolinguistic research on linguistic innovation, variation, and change.

Fibro-CoSANet: pulmonary fibrosis prognosis prediction using a convolutional self attention network

Idiopathic pulmonary fibrosis (IPF) is a restrictive interstitial lung disease that causes lung function decline by lung tissue scarring. Although lung function decline is assessed by the forced vital capacity (FVC), determining the accurate progression of IPF remains a challenge. To address this challenge, we proposed Fibro-CoSANet, a novel end-to-end multi-modal learning based approach, to predict the FVC decline. Fibro-CoSANet utilized computed tomography images and demographic information in convolutional neural network frameworks with a stacked attention layer. Extensive experiments on the OSIC Pulmonary Fibrosis Progression Dataset demonstrated the superiority of our proposed Fibro-CoSANet by achieving new state-of-the-art modified Laplace log-likelihood score of -6.68. This network may benefit research areas concerned with designing networks to improve the prognostic accuracy of IPF. The source-code for Fibro-CoSANet is available at: https://github.com/zabir-nabil/Fibro-CoSANet.

Deep learning-a first meta-survey of selected reviews across scientific disciplines, their commonalities, challenges and research impact

Deep learning belongs to the field of artificial intelligence, where machines perform tasks that typically require some kind of human intelligence. Deep learning tries to achieve this by drawing inspiration from the learning of a human brain. Similar to the basic structure of a brain, which consists of (billions of) neurons and connections between them, a deep learning algorithm consists of an artificial neural network, which resembles the biological brain structure. Mimicking the learning process of humans with their senses, deep learning networks are fed with (sensory) data, like texts, images, videos or sounds. These networks outperform the state-of-the-art methods in different tasks and, because of this, the whole field saw an exponential growth during the last years. This growth resulted in way over 10,000 publications per year in the last years. For example, the search engine PubMed alone, which covers only a sub-set of all publications in the medical field, provides already over 11,000 results in Q3 2020 for the search term 'deep learning', and around 90% of these results are from the last three years. Consequently, a complete overview over the field of deep learning is already impossible to obtain and, in the near future, it will potentially become difficult to obtain an overview over a subfield. However, there are several review articles about deep learning, which are focused on specific scientific fields or applications, for example deep learning advances in computer vision or in specific tasks like object detection. With these surveys as a foundation, the aim of this contribution is to provide a first high-level, categorized meta-survey of selected reviews on deep learning across different scientific disciplines and outline the research impact that they already have during a short period of time. The categories (computer vision, language processing, medical informatics and additional works) have been chosen according to the underlying data sources (image, language, medical, mixed). In addition, we review the common architectures, methods, pros, cons, evaluations, challenges and future directions for every sub-category.

QF-TraderNet: Intraday Trading via Deep Reinforcement With Quantum Price Levels Based Profit-And-Loss Control

Reinforcement Learning (RL) based machine trading attracts a rich profusion of interest. However, in the existing research, RL in the day-trade task suffers from the noisy financial movement in the short time scale, difficulty in order settlement, and expensive action search in a continuous-value space. This paper introduced an end-to-end RL intraday trading agent, namely QF-TraderNet, based on the quantum finance theory (QFT) and deep reinforcement learning. We proposed a novel design for the intraday RL trader’s action space, inspired by the Quantum Price Levels (QPLs). Our action space design also brings the model a learnable profit-and-loss control strategy. QF-TraderNet composes two neural networks: 1) A long short term memory networks for the feature learning of financial time series; 2) a policy generator network (PGN) for generating the distribution of actions. The profitability and robustness of QF-TraderNet have been verified in multi-type financial datasets, including FOREX, metals, crude oil, and financial indices. The experimental results demonstrate that QF-TraderNet outperforms other baselines in terms of cumulative price returns and Sharpe Ratio, and the robustness in the acceidential market shift.

Machine Learning for Recurrence Prediction of Gynecologic Cancers Using Lynch Syndrome-Related Screening Markers

Simple Summary

Recurrent patients with gynecologic cancer experience a difficult situation when using immune checkpoint inhibitors based on mismatch repair gene immunohistochemistry and microsatellite instability. Six machine learning algorithms were used to create predictive models with seven prospective features (four MMR IHC [MLH1, MSH2, MSH6, and PMS2], MSI, Age 60, and tumor size). The experimental results showed that the RF-based model performed best at predicting gynecologic cancer recurrence, with AUCs of 0.818 and 0.826 for the 5-fold cross-validation (CV) and 5-fold CV with 10 repetitions, respectively. This provides novel and baseline results of patients with recurrent gynecologic cancer using immune checkpoint inhibitors by using machine learning methods based on Lynch syndrome-related screening markers.

Abstract

To support the implementation of genome-based precision medicine, we developed machine learning models that predict the recurrence of patients with gynecologic cancer in using immune checkpoint inhibitors (ICI) based on clinical and pathologic characteristics, including Lynch syndrome-related screening markers such as immunohistochemistry (IHC) and microsatellite instability (MSI) tests. To accomplish our goal, we reviewed the patient demographics, clinical data, and pathological results from their medical records. Then we identified seven potential characteristics (four MMR IHC [MLH1, MSH2, MSH6, and PMS2], MSI, Age 60, and tumor size). Following that, predictive models were built based on these variables using six machine learning algorithms: logistic regression (LR), support vector machine (SVM), naive Bayes (NB), random forest (RF), gradient boosting (GB), and extreme gradient boosting (EGB) (XGBoost). The experimental results showed that the RF-based model performed best at predicting gynecologic cancer recurrence, with AUCs of 0.818 and 0.826 for the 5-fold cross-validation (CV) and 5-fold CV with 10 repetitions, respectively. This study provides novel and baseline results about predicting the recurrence of gynecologic cancer in patients using ICI by using machine learning methods based on Lynch syndrome-related screening markers.

Web-based snake identification service: A successful model of snake identification in Sri Lanka

Snakes are reptiles of great biomedical significance. The accurate identification of snakes is particularly important for healthcare workers to diagnose and treat victims of snakebite envenoming. Further, snake identification is vital for the general population, especially to those who live in areas of high snakebite incidence. Owing to the great diversity of snakes and the superficial similarities between some species, the correct identification of these reptiles is often difficult. Therefore, identification of snake species is challenging for healthcare workers, biologists, naturalists, and the general population. To overcome this challenge, we developed a web-based snake identification service (www.snakesidentification.org) in Sri Lanka, which provides rapid and accurate identification by experienced herpetologists. This service received 486 identification requests over a period of 40 months. The majority of requests were from Colombo District [140 (28.8%)], though only 63 (13.0%) of these were identified as medically important snakes. The majority [389 (80.0%)] of the requests related either to feebly venomous colubrid snakes or non-venomous species. The sample included 30 (of 107) snake species in the island, including 8 endemic species. There were 315 (64.8%) requests relating to live snakes. In the majority of cases (285, 90.4%), the snake was released to the closest available habitat after being identified. The median time taken to respond to requests was 70 min (interquartile range 23-299 min). The majority of persons making requests (283, 58.2%) were unable to identify the snakes. For those who attempted identification the snakes, correct identification was made by only 59 (12.1%), whereas 144 (29.6%) identified the snake incorrectly. This web-based snake identification service provides an example of a successful and useful model of rapid snake identification. Similar models could be implemented in other regions and countries to provide accurate information on snake identification both to the healthcare workers and the general public.

Decoding the molecular subtypes of breast cancer seen on multimodal ultrasound images using an assembled convolutional neural network model: A prospective and multicentre study

Background

Preoperative determination of breast cancer molecular subtypes facilitates individualized treatment plan-making and improves patient prognosis. We aimed to develop an assembled convolutional neural network (ACNN) model for the preoperative prediction of molecular subtypes using multimodal ultrasound (US) images.

Methods

This multicentre study prospectively evaluated a dataset of greyscale US, colour Doppler flow imaging (CDFI), and shear-wave elastography (SWE) images in 807 patients with 818 breast cancers from November 2016 to February 2021. The St. Gallen molecular subtypes of breast cancer were confirmed by postoperative immunohistochemical examination. The monomodal ACNN model based on greyscale US images, the dual-modal ACNN model based on greyscale US and CDFI images, and the multimodal ACNN model based on greyscale US and CDFI as well as SWE images were constructed in the training cohort. The performances of three ACNN models in predicting four- and five-classification molecular subtypes and identifying triple negative from non-triple negative subtypes were assessed and compared. The performance of the multimodal ACNN was also compared with preoperative core needle biopsy (CNB).

Finding

The performance of the multimodal ACNN model (macroaverage area under the curve [AUC]: 0.89–0.96) was superior to that of the dual-modal ACNN model (macroaverage AUC: 0.81–0.84) and the monomodal ACNN model (macroaverage AUC: 0.73–0.75) in predicting four-classification breast cancer molecular subtypes, which was also better than that of preoperative CNB (AUC: 0.89–0.99 vs. 0.67–0.82, p < 0.05). In addition, the multimodal ACNN model outperformed the other two ACNN models in predicting five-classification molecular subtypes (AUC: 0.87–0.94 vs. 0.78-0.81 vs. 0.71–0.78) and identifying triple negative from non-triple negative breast cancers (AUC: 0.934–0.970 vs. 0.688–0.830 vs. 0.536–0.650, p < 0.05). Moreover, the multimodal ACNN model obtained satisfactory prediction performance for both T1 and non-T1 lesions (AUC: 0.957–0.958 and 0.932–0.985).

Interpretation

The multimodal US-based ACNN model is a potential noninvasive decision-making method for the management of patients with breast cancer in clinical practice.

Funding

This work was supported in part by the National Natural Science Foundation of China (Grants 81725008 and 81927801), Shanghai Municipal Health Commission (Grants 2019LJ21 and SHSLCZDZK03502), and the Science and Technology Commission of Shanghai Municipality (Grants 19441903200, 19DZ2251100, and 21Y11910800).

Infoecology of the deep learning and smart manufacturing: thematic and concept interactions

Purpose

This infecological study mainly aimed to know the thematic and conceptual relationship in published papers in deep learning (DL) and smart manufacturing (SM).

Design/methodology/approach

The research methodology has specific research objectives based on the type and method of research, data analysis tools. In general, description methods are applied by Web of Science (WoS) analysis tools and Voyant tools as a web-based reading and analysis environment for digital texts. The Yewno tool is applied to draw a knowledge map to show the concept's interaction between DL and SM.

Findings

The knowledge map of DL and SM concepts shows that there are currently few concepts interacting with each other, while the rapid growth of technology and the needs of today's society have revealed the need to use more and more DL in SM. The results of this study can provide a coherent and well-mapped road map to the main policymakers of the field of research in DL and SM, through the study of coexistence and interactions of the thematic categories with other thematic areas. In this way, they can design more effective guidelines and strategies to solve the problems of researchers in conducting their studies and direct. The analysis results demonstrated that the information ecosystem of DL and SM studies dynamically developed over time. The continuous conduction flow of scientific studies in this field brought continuous changes into the infoecology of subjects and concepts in this area.

Originality/value

The paper investigated the thematic interaction of the scientific productions in DL and SM and discussed possible implications. We used of the variety tools and techniques to draw our own perspective. Also, we drew arguments from other research work to back up our findings.

Structural Insights on the SARS-CoV-2 Variants of Concern Spike Glycoprotein: A Computational Study With Possible Clinical Implications

Coronavirus disease 2019 (COVID-19) pandemic has been attributed to SARS-CoV-2 (SARS2) and, consequently, SARS2 has evolved into multiple SARS2 variants driving subsequent waves of infections. In particular, variants of concern (VOC) were identified to have both increased transmissibility and virulence ascribable to mutational changes occurring within the spike protein resulting to modifications in the protein structural orientation which in-turn may affect viral pathogenesis. However, this was never fully elucidated. Here, we generated spike models of endemic HCoVs (HCoV 229E, HCoV OC43, HCoV NL63, HCoV HKU1, SARS CoV, MERS CoV), original SARS2, and VOC (alpha, beta, gamma, delta). Model quality check, structural superimposition, and structural comparison based on RMSD values, TM scores, and contact mapping were all performed. We found that: 1) structural comparison between the original SARS2 and VOC whole spike protein model have minor structural differences (TM > 0.98); 2) the whole VOC spike models putatively have higher structural similarity (TM > 0.70) to spike models from endemic HCoVs coming from the same phylogenetic cluster; 3) original SARS2 S1-CTD and S1-NTD models are structurally comparable to VOC S1-CTD (TM = 1.0) and S1-NTD (TM > 0.96); and 4) endemic HCoV S1-CTD and S1-NTD models are structurally comparable to VOC S1-CTD (TM > 0.70) and S1-NTD (TM > 0.70) models belonging to the same phylogenetic cluster. Overall, we propose that structural similarities (possibly ascribable to similar conformational epitopes) may help determine immune cross-reactivity, whereas, structural differences (possibly associated with varying conformational epitopes) may lead to viral infection (either reinfection or breakthrough infection).

Brain tumor detection and classification using machine learning: a comprehensive survey

Brain tumor occurs owing to uncontrolled and rapid growth of cells. If not treated at an initial phase, it may lead to death. Despite many significant efforts and promising outcomes in this domain, accurate segmentation and classification remain a challenging task. A major challenge for brain tumor detection arises from the variations in tumor location, shape, and size. The objective of this survey is to deliver a comprehensive literature on brain tumor detection through magnetic resonance imaging to help the researchers. This survey covered the anatomy of brain tumors, publicly available datasets, enhancement techniques, segmentation, feature extraction, classification, and deep learning, transfer learning and quantum machine learning for brain tumors analysis. Finally, this survey provides all important literature for the detection of brain tumors with their advantages, limitations, developments, and future trends.

Deep Learning-based Sentiment Analysis and Topic Modeling on Tourism During Covid-19 Pandemic

The Covid-19 pandemic has disrupted the world economy and significantly influenced the tourism industry. Millions of people have shared their emotions, views, facts, and circumstances on numerous social media platforms, which has resulted in a massive flow of information. The high-density social media data has drawn many researchers to extract valuable information and understand the user’s emotions during the pandemic time. The research looks at the data collected from the micro-blogging site Twitter for the tourism sector, emphasizing sub-domains hospitality and healthcare. The sentiment of approximately 20,000 tweets have been calculated using Valence Aware Dictionary for Sentiment Reasoning (VADER) model. Furthermore, topic modeling was used to reveal certain hidden themes and determine the narrative and direction of the topics related to tourism healthcare, and hospitality. Topic modeling also helped us to identify inter-cluster similar terms and analyzing the flow of information from a group of a similar opinion. Finally, a cutting-edge deep learning classification model was used with different epoch sizes of the dataset to anticipate and classify the people’s feelings. The deep learning model has been tested with multiple parameters such as training set accuracy, test set accuracy, validation loss, validation accuracy, etc., and resulted in more than a 90% in training set accuracy tourism hospitality and healthcare reported 80.9 and 78.7% respectively on test set accuracy.

Understanding, Explanation, and Active Inference

While machine learning techniques have been transformative in solving a range of problems, an important challenge is to understand why they arrive at the decisions they output. Some have argued that this necessitates augmenting machine intelligence with understanding such that, when queried, a machine is able to explain its behaviour (i.e., explainable AI). In this article, we address the issue of machine understanding from the perspective of active inference. This paradigm enables decision making based upon a model of how data are generated. The generative model contains those variables required to explain sensory data, and its inversion may be seen as an attempt to explain the causes of these data. Here we are interested in explanations of one's own actions. This implies a deep generative model that includes a model of the world, used to infer policies, and a higher-level model that attempts to predict which policies will be selected based upon a space of hypothetical (i.e., counterfactual) explanations-and which can subsequently be used to provide (retrospective) explanations about the policies pursued. We illustrate the construct validity of this notion of understanding in relation to human understanding by highlighting the similarities in computational architecture and the consequences of its dysfunction.

The perspective of IT decision makers on factors influencing adoption and implementation of AI-technologies in 40 German Hospitals: Descriptive Analysis (Preprint)

Background

New artificial intelligence (AI) tools are being developed at a high speed. However, strategies and practical experiences surrounding the adoption and implementation of AI in health care are lacking. This is likely because of the high implementation complexity of AI, legacy IT infrastructure, and unclear business cases, thus complicating AI adoption. Research has recently started to identify the factors influencing AI readiness of organizations.

Objective

This study aimed to investigate the factors influencing AI readiness as well as possible barriers to AI adoption and implementation in German hospitals. We also assessed the status quo regarding the dissemination of AI tools in hospitals. We focused on IT decision makers, a seldom studied but highly relevant group.

Methods

We created a web-based survey based on recent AI readiness and implementation literature. Participants were identified through a publicly accessible database and contacted via email or invitational leaflets sent by mail, in some cases accompanied by a telephonic prenotification. The survey responses were analyzed using descriptive statistics.

Results

We contacted 609 possible participants, and our database recorded 40 completed surveys. Most participants agreed or rather agreed with the statement that AI would be relevant in the future, both in Germany (37/40, 93%) and in their own hospital (36/40, 90%). Participants were asked whether their hospitals used or planned to use AI technologies. Of the 40 participants, 26 (65%) answered “yes.” Most AI technologies were used or planned for patient care, followed by biomedical research, administration, and logistics and central purchasing. The most important barriers to AI were lack of resources (staff, knowledge, and financial). Relevant possible opportunities for using AI were increase in efficiency owing to time-saving effects, competitive advantages, and increase in quality of care. Most AI tools in use or in planning have been developed with external partners.

Conclusions

Few tools have been implemented in routine care, and many hospitals do not use or plan to use AI in the future. This can likely be explained by missing or unclear business cases or the need for a modern IT infrastructure to integrate AI tools in a usable manner. These shortcomings complicate decision-making and resource attribution. As most AI technologies already in use were developed in cooperation with external partners, these relationships should be fostered. IT decision makers should assess their hospitals’ readiness for AI individually with a focus on resources. Further research should continue to monitor the dissemination of AI tools and readiness factors to determine whether improvements can be made over time. This monitoring is especially important with regard to government-supported investments in AI technologies that could alleviate financial burdens. Qualitative studies with hospital IT decision makers should be conducted to further explore the reasons for slow AI.

Computational methods directed towards drug repurposing for COVID-19: advantages and limitations

Novel coronavirus disease 2019 (COVID-19) has significantly altered the socio-economic status of countries. Although vaccines are now available against the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), a causative agent for COVID-19, it continues to transmit and newer variants of concern have been consistently emerging world-wide. Computational strategies involving drug repurposing offer a viable opportunity to choose a medication from a rundown of affirmed drugs against distinct diseases including COVID-19. While pandemics impede the healthcare systems, drug repurposing or repositioning represents a hopeful approach in which existing drugs can be remodeled and employed to treat newer diseases. In this review, we summarize the diverse computational approaches attempted for developing drugs through drug repurposing or repositioning against COVID-19 and discuss their advantages and limitations. To this end, we have outlined studies that utilized computational techniques such as molecular docking, molecular dynamic simulation, disease-disease association, drug-drug interaction, integrated biological network, artificial intelligence, machine learning and network medicine to accelerate creation of smart and safe drugs against COVID-19.

The Impact of Explanations on Layperson Trust in Artificial Intelligence-Driven Symptom Checker Apps: Experimental Study

BACKGROUND

Artificial intelligence (AI)-driven symptom checkers are available to millions of users globally and are advocated as a tool to deliver health care more efficiently. To achieve the promoted benefits of a symptom checker, laypeople must trust and subsequently follow its instructions. In AI, explanations are seen as a tool to communicate the rationale behind black-box decisions to encourage trust and adoption. However, the effectiveness of the types of explanations used in AI-driven symptom checkers has not yet been studied. Explanations can follow many forms, including why-explanations and how-explanations. Social theories suggest that why-explanations are better at communicating knowledge and cultivating trust among laypeople.

OBJECTIVE

The aim of this study is to ascertain whether explanations provided by a symptom checker affect explanatory trust among laypeople and whether this trust is impacted by their existing knowledge of disease.

METHODS

A cross-sectional survey of 750 healthy participants was conducted. The participants were shown a video of a chatbot simulation that resulted in the diagnosis of either a migraine or temporal arteritis, chosen for their differing levels of epidemiological prevalence. These diagnoses were accompanied by one of four types of explanations. Each explanation type was selected either because of its current use in symptom checkers or because it was informed by theories of contrastive explanation. Exploratory factor analysis of participants' responses followed by comparison-of-means tests were used to evaluate group differences in trust.

RESULTS

Depending on the treatment group, two or three variables were generated, reflecting the prior knowledge and subsequent mental model that the participants held. When varying explanation type by disease, migraine was found to be nonsignificant (P=.65) and temporal arteritis, marginally significant (P=.09). Varying disease by explanation type resulted in statistical significance for input influence (P=.001), social proof (P=.049), and no explanation (P=.006), with counterfactual explanation (P=.053). The results suggest that trust in explanations is significantly affected by the disease being explained. When laypeople have existing knowledge of a disease, explanations have little impact on trust. Where the need for information is greater, different explanation types engender significantly different levels of trust. These results indicate that to be successful, symptom checkers need to tailor explanations to each user's specific question and discount the diseases that they may also be aware of.

CONCLUSIONS

System builders developing explanations for symptom-checking apps should consider the recipient's knowledge of a disease and tailor explanations to each user's specific need. Effort should be placed on generating explanations that are personalized to each user of a symptom checker to fully discount the diseases that they may be aware of and to close their information gap.

Data-driven operation of the resilient electric grid: A case of COVID-19

Electrical energy is a vital part of modern life, and expectations for grid resilience to allow a continuous and reliable energy supply has tremendously increased even during adverse events (e.g. Ukraine cyberattack, Hurricane Maria). The global pandemic COVID-19 has raised the electric energy reliability risk due to potential workforce disruptions, supply chain interruptions, and increased possible cybersecurity threats. Additionally, the pandemic introduces a significant degree of uncertainty to the grid operation in the presence of other challenges including aging power grids, high proliferation of distributed generation, market mechanism, and active distribution network. This situation increases the need for measures for the resiliency of power grids to mitigate the impact of the pandemic as well as simultaneous extreme events including cyberattacks and adverse weather events. Solutions to manage such an adverse scenario will be multi-fold: (a) emergency planning and organisational support, (b) following safety protocol, (c) utilising enhanced automation and sensing for situational awareness, and (d) integration of advanced technologies and data points for ML-driven enhanced decision support. Enhanced digitalisation and automation resulted in better network visibility at various levels, including generation, transmission, and distribution. These data or information can be employed to take advantage of advanced machine learning techniques for automation and increased power grid resilience. In this paper, the resilience of power grids in the face of pandemics is explored and various machine learning tools that can be helpful to augment human operators are discused by: (a) reviewing the impact of COVID-19 on power grid operations and actions taken by operators/organisations to minimise the impact of COVID-19, and (b) presenting recently developed tools and concepts of machine learning and artificial intelligence that can be applied to increase the resiliency of power systems in normal and extreme scenarios such as the COVID-19 pandemic.

Automated Cerebral Vessel Segmentation of Magnetic Resonance Imaging in Patients with Intracranial Atherosclerotic Diseases

Time-of-flight (TOF) magnetic resonance angiography is a non-invasive imaging modality for the diagnosis of intracranial atherosclerotic diseases (ICAD). Evaluation of the degree of the stenosis and status of posterior and anterior communicating arteries to supply enough blood flow to the distal arteries is very critical, which requires accurate evaluation of arteries. Recently, deep-learning methods have been firmly established as a robust tool in medical image segmentation, which has been resulted in developing multiple customized algorithms. For instance, BRAVE-NET, a context-based successor of U-Net-has shown promising results in MRA cerebrovascular segmentation. Another widely used context-based 3D CNN-DeepMedic-has been shown to outperform U-Net in cerebrovascular segmentation of 3D digital subtraction angiography. In this study, we aim to train and compare the two state-of-the-art deep-learning networks, BRAVE-NET and DeepMedic, for automated and reliable brain vessel segmentation from TOF-MRA images in ICAD patients. Using specially labeled data-labeled on TOF MRA and corrected on high-resolution black-blood MRI, of 51 patients with ICAD due to severe stenosis, we trained and tested both models. On an independent test dataset of 11 cases, DeepMedic slightly outperformed BRAVE-NET in terms of DSC (0.905±0.012 vs 0.893±0.015, p: 0.539) and 95HD (0.754±0.223 vs 1.768±0.609, p: 0.134), and significantly outperformed BRAVE-NET in terms of Recall (0.940±0.023 vs 0.855±0.030, p: 0.036). Qualitative assessment confirmed the superiority of DeepMedic in capturing the small and distal arteries. While BRAVE-NET consistently reported higher precision, DeepMedic generally overpredicted and could better visualize the smaller and distal arteries. In future studies, ensemble models that can leverage best of both should be developed and tested on larger datasets.Clinical Relevance- This study helps elevate the state-of-the-art for brain vessel segmentation from non-invasive MRA, which could accelerate the translation of vessel status-based biomarkers into the clinical setting.