Prospective validation of a seizure diary forecasting falls short.

OBJECTIVE: Recently, a deep learning artificial intelligence (AI) model forecasted seizure risk using retrospective seizure diaries with higher accuracy than random forecasts. The present study sought to prospectively evaluate the same algorithm. METHODS: We recruited a prospective cohort of 46 people with epilepsy; 25 completed sufficient data entry for analysis (median = 5 months). We used the same AI method as in our prior study. Group-level and individual-level Brier Skill Scores (BSSs) compared random forecasts and simple moving average forecasts to the AI. RESULTS: The AI had an area under the receiver operating characteristic curve of .82. At the group level, the AI outperformed random forecasting (BSS = .53). At the individual level, AI outperformed random in 28% of cases. At the group and individual level, the moving average outperformed the AI. If pre-enrollment (nonverified) diaries (with presumed underreporting) were included, the AI significantly outperformed both comparators. Surveys showed most did not mind poor-quality LOW-RISK or HIGH-RISK forecasts, yet 91% wanted access to these forecasts. SIGNIFICANCE: The previously developed AI forecasting tool did not outperform a very simple moving average forecasting in this prospective cohort, suggesting that the AI model should be replaced.

Deep Learning in Cancer Diagnosis and Prognosis Prediction: A Minireview on Challenges, Recent Trends, and Future Directions.

Deep learning (DL) is a branch of machine learning and artificial intelligence that has been applied to many areas in different domains such as health care and drug design. Cancer prognosis estimates the ultimate fate of a cancer subject and provides survival estimation of the subjects. An accurate and timely diagnostic and prognostic decision will greatly benefit cancer subjects. DL has emerged as a technology of choice due to the availability of high computational resources. The main components in a standard computer-aided design (CAD) system are preprocessing, feature recognition, extraction and selection, categorization, and performance assessment. Reduction of costs associated with sequencing systems offers a myriad of opportunities for building precise models for cancer diagnosis and prognosis prediction. In this survey, we provided a summary of current works where DL has helped to determine the best models for the cancer diagnosis and prognosis prediction tasks. DL is a generic model requiring minimal data manipulations and achieves better results while working with enormous volumes of data. Aims are to scrutinize the influence of DL systems using histopathology images, present a summary of state-of-the-art DL methods, and give directions to future researchers to refine the existing methods.

New Insights Into Drug Repurposing for COVID-19 Using Deep Learning.

The coronavirus disease 2019 (COVID-19) has continued to spread worldwide since late 2019. To expedite the process of providing treatment to those who have contracted the disease and to ensure the accessibility of effective drugs, numerous strategies have been implemented to find potential anti-COVID-19 drugs in a short span of time. Motivated by this critical global challenge, in this review, we detail approaches that have been used for drug repurposing for COVID-19 and suggest improvements to the existing deep learning (DL) approach to identify and repurpose drugs to treat this complex disease. By optimizing hyperparameter settings, deploying suitable activation functions, and designing optimization algorithms, the improved DL approach will be able to perform feature extraction from quality big data, turning the traditional DL approach, referred to as a "black box," which generalizes and learns the transmitted data, into a "glass box" that will have the interpretability of its rationale while maintaining a high level of prediction accuracy. When adopted for drug repurposing for COVID-19, this improved approach will create a new generation of DL approaches that can establish a cause and effect relationship as to why the repurposed drugs are suitable for treating COVID-19. Its ability can also be extended to repurpose drugs for other complex diseases, develop appropriate treatment strategies for new diseases, and provide precision medical treatment to patients, thus paving the way to discover new drugs that can potentially be effective for treating COVID-19.

Deep learning assisted quantitative assessment of histopathological markers of Alzheimer's disease and cerebral amyloid angiopathy.

Traditionally, analysis of neuropathological markers in neurodegenerative diseases has relied on visual assessments of stained sections. Resulting semiquantitative scores often vary between individual raters and research centers, limiting statistical approaches. To overcome these issues, we have developed six deep learning-based models, that identify some of the most characteristic markers of Alzheimer's disease (AD) and cerebral amyloid angiopathy (CAA). The deep learning-based models are trained to differentially detect parenchymal amyloid ß (Aß)-plaques, vascular Aß-deposition, iron and calcium deposition, reactive astrocytes, microglia, as well as fibrin extravasation. The models were trained on digitized histopathological slides from brains of patients with AD and CAA, using a workflow that allows neuropathology experts to train convolutional neural networks (CNNs) on a cloud-based graphical interface. Validation of all models indicated a very good to excellent performance compared to three independent expert human raters. Furthermore, the Aß and iron models were consistent with previously acquired semiquantitative scores in the same dataset and allowed the use of more complex statistical approaches. For example, linear mixed effects models could be used to confirm the previously described relationship between leptomeningeal CAA severity and cortical iron accumulation. A similar approach enabled us to explore the association between neuroinflammation and disparate Aß pathologies. The presented workflow is easy for researchers with pathological expertise to implement and is customizable for additional histopathological markers. The implementation of deep learning-assisted analyses of histopathological slides is likely to promote standardization of the assessment of neuropathological markers across research centers, which will allow specific pathophysiological questions in neurodegenerative disease to be addressed in a harmonized way and on a larger scale.

Deep learning-based natural language processing in ophthalmology: applications, challenges and future directions.

PURPOSE OF REVIEW: Artificial intelligence (AI) is the fourth industrial revolution in mankind's history. Natural language processing (NLP) is a type of AI that transforms human language, to one that computers can interpret and process. NLP is still in the formative stages of development in healthcare, with promising applications and potential challenges in its applications. This review provides an overview of AI-based NLP, its applications in healthcare and ophthalmology, next-generation use case, as well as potential challenges in deployment. RECENT FINDINGS: The integration of AI-based NLP systems into existing clinical care shows considerable promise in disease screening, risk stratification, and treatment monitoring, amongst others. Stakeholder collaboration, greater public acceptance, and advancing technologies will continue to shape the NLP landscape in healthcare and ophthalmology. SUMMARY: Healthcare has always endeavored to be patient centric and personalized. For AI-based NLP systems to become an eventual reality in larger-scale applications, it is pertinent for key stakeholders to collaborate and address potential challenges in application. Ultimately, these would enable more equitable and generalizable use of NLP systems for the betterment of healthcare and society.

AIDeveloper: Deep Learning Image Classification in Life Science and Beyond.

Artificial intelligence (AI)-based image analysis has increased drastically in recent years. However, all applications use individual solutions, highly specialized for a particular task. Here, an easy-to-use, adaptable, and open source software, called AIDeveloper (AID) to train neural nets (NN) for image classification without the need for programming is presented. AID provides a variety of NN-architectures, allowing to apply trained models on new data, obtain performance metrics, and export final models to different formats. AID is benchmarked on large image datasets (CIFAR-10 and Fashion-MNIST). Furthermore, models are trained to distinguish areas of differentiated stem cells in images of cell culture. A conventional blood cell count and a blood count obtained using an NN are compared, trained on >1.2 million images, and demonstrated how AID can be used for label-free classification of B- and T-cells. All models are generated by non-programmers on generic computers, allowing for an interdisciplinary use.

Predicting innovative firms using web mining and deep learning.

Evidence-based STI (science, technology, and innovation) policy making requires accurate indicators of innovation in order to promote economic growth. However, traditional indicators from patents and questionnaire-based surveys often lack coverage, granularity as well as timeliness and may involve high data collection costs, especially when conducted at a large scale. Consequently, they struggle to provide policy makers and scientists with the full picture of the current state of the innovation system. In this paper, we propose a first approach on generating web-based innovation indicators which may have the potential to overcome some of the shortcomings of traditional indicators. Specifically, we develop a method to identify product innovator firms at a large scale and very low costs. We use traditional firm-level indicators from a questionnaire-based innovation survey (German Community Innovation Survey) to train an artificial neural network classification model on labelled (product innovator/no product innovator) web texts of surveyed firms. Subsequently, we apply this classification model to the web texts of hundreds of thousands of firms in Germany to predict whether they are product innovators or not. We then compare these predictions to firm-level patent statistics, survey extrapolation benchmark data, and regional innovation indicators. The results show that our approach produces reliable predictions and has the potential to be a valuable and highly cost-efficient addition to the existing set of innovation indicators, especially due to its coverage and regional granularity.

A generative adversarial network approach to (ensemble) weather prediction.

We use a conditional deep convolutional generative adversarial network to predict the geopotential height of the 500 hPa pressure level, the two-meter temperature and the total precipitation for the next 24 h over Europe. The proposed models are trained on 4 years of ERA5 reanalysis data from with the goal to predict the associated meteorological fields in 2019. The forecasts show a good qualitative and quantitative agreement with the true reanalysis data for the geopotential height and two-meter temperature, while failing for total precipitation, thus indicating that weather forecasts based on data alone may be possible for specific meteorological parameters. We further use Monte-Carlo dropout to develop an ensemble weather prediction system based purely on deep learning strategies, which is computationally cheap and further improves the skill of the forecasting model, by allowing to quantify the uncertainty in the current weather forecast as learned by the model.

Deep Learning Application in Spinal Implant Identification.

STUDY DESIGN: Retrospective observational study. OBJECTIVE: To demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning. SUMMARY OF BACKGROUND DATA: Deep learning has recently been actively applied to medical images. However, despite many attempts to apply deep learning to medical images, the application has rarely been successful. We aimed to demonstrate the effectiveness and usefulness of deep learning in the medical field. The goal of this study was to demonstrate the clinical usefulness of deep learning by identifying previous spinal implants through application of deep learning. METHODS: For deep learning algorithm development, radiographs were retrospectively obtained from clinical cases in which the patients had lumbar spine one-segment instrument surgery. A total of 2894 lumbar spine anteroposterior (AP: 1446 cases) and lateral (1448 cases) radiographs were collected. Labeling work was conducted for five different implants. We conducted experiments using three deep learning algorithms. The traditional deep neural network model built by coding the transfer learning algorithm, Google AutoML, and Apple Create ML. Recall (sensitivity) and precision (specificity) were measured after training. RESULTS: Overall, each model performed well in identifying each pedicle screw implant. In conventional transfer learning, AP radiography showed 97.0% precision and 96.7% recall. Lateral radiography showed 98.7% precision and 98.2% recall. In Google AutoML, AP radiography showed 91.4% precision and 87.4% recall; lateral radiography showed 97.9% precision and 98.4% recall. In Apple Create ML, AP radiography showed 76.0% precision and 73.0% recall; lateral radiography showed 89.0% precision and 87.0% recall. In all deep learning algorithms, precision and recall were higher in lateral than in AP radiography. CONCLUSION: The deep learning application is effective for spinal implant identification. This demonstrates that clinicians can use ML-based deep learning applications to improve clinical practice and patient care.Level of Evidence: 3.

Deep learning models for the prediction of intraoperative hypotension.

BACKGROUND: Intraoperative hypotension is associated with a risk of postoperative organ dysfunction. In this study, we aimed to present deep learning algorithms for real-time predictions 5, 10, and 15 min before a hypotensive event. METHODS: In this retrospective observational study, deep learning algorithms were developed and validated using biosignal waveforms acquired from patient monitoring of noncardiac surgery. The classification model was a binary classifier of a hypotensive event (MAP <65 mm Hg) or a non-hypotensive event by analysing biosignal waveforms. The regression model was developed to directly estimate the MAP. The primary outcome was area under the receiver operating characteristic (AUROC) curve and the mean absolute error (MAE). RESULTS: In total, 3301 patients were included. For invasive models, the multichannel model with an arterial pressure waveform, electrocardiography, photoplethysmography, and capnography showed greater AUROC than the arterial-pressure-only models (AUROC15-min, 0.897 [95% confidence interval {CI}: 0.894-0.900] vs 0.891 [95% CI: 0.888-0.894]) and lesser MAE (MAE15-min, 7.76 mm Hg [95% CI: 7.64-7.87 mm Hg] vs 8.12 mm Hg [95% CI: 8.02-8.21 mm Hg]). For the noninvasive models, the multichannel model showed greater AUROCs than that of the photoplethysmography-only models (AUROC15-min, 0.762 [95% CI: 0.756-0.767] vs 0.694 [95% CI: 0.686-0.702]) and lesser MAEs (MAE15-min, 11.68 mm Hg [95% CI: 11.57-11.80 mm Hg] vs 12.67 [95% CI: 12.56-12.79 mm Hg]). CONCLUSIONS: Deep learning models can predict hypotensive events based on biosignals acquired using invasive and noninvasive patient monitoring. In addition, the model shows better performance when using combined rather than single signals.

t-soft update of target network for deep reinforcement learning.

This paper proposes a new robust update rule of target network for deep reinforcement learning (DRL), to replace the conventional update rule, given as an exponential moving average. The target network is for smoothly generating the reference signals for a main network in DRL, thereby reducing learning variance. The problem with its conventional update rule is the fact that all the parameters are smoothly copied with the same speed from the main network, even when some of them are trying to update toward the wrong directions. This behavior increases the risk of generating the wrong reference signals. Although slowing down the overall update speed is a naive way to mitigate wrong updates, it would decrease learning speed. To robustly update the parameters while keeping learning speed, a t-soft update method, which is inspired by Student-t distribution, is derived with reference to the analogy between the exponential moving average and the normal distribution. Through the analysis of the derived t-soft update, we show that it takes over the properties of the Student-t distribution. Specifically, with a heavy-tailed property of the Student-t distribution, the t-soft update automatically excludes extreme updates that differ from past experiences. In addition, when the updates are similar to the past experiences, it can mitigate the learning delay by increasing the amount of updates. In PyBullet robotics simulations for DRL, an online actor-critic algorithm with the t-soft update outperformed the conventional methods in terms of the obtained return and/or its variance. From the training process by the t-soft update, we found that the t-soft update is globally consistent with the standard soft update, and the update rates are locally adjusted for acceleration or suppression.

Machine and Deep Learning towards COVID-19 Diagnosis and Treatment: Survey, Challenges, and Future Directions.

With many successful stories, machine learning (ML) and deep learning (DL) have been widely used in our everyday lives in a number of ways. They have also been instrumental in tackling the outbreak of Coronavirus (COVID-19), which has been happening around the world. The SARS-CoV-2 virus-induced COVID-19 epidemic has spread rapidly across the world, leading to international outbreaks. The COVID-19 fight to curb the spread of the disease involves most states, companies, and scientific research institutions. In this research, we look at the Artificial Intelligence (AI)-based ML and DL methods for COVID-19 diagnosis and treatment. Furthermore, in the battle against COVID-19, we summarize the AI-based ML and DL methods and the available datasets, tools, and performance. This survey offers a detailed overview of the existing state-of-the-art methodologies for ML and DL researchers and the wider health community with descriptions of how ML and DL and data can improve the status of COVID-19, and more studies in order to avoid the outbreak of COVID-19. Details of challenges and future directions are also provided.

Review of deep learning algorithms for the automatic detection of intracranial hemorrhages on computed tomography head imaging.

Artificial intelligence is a rapidly evolving field, with modern technological advances and the growth of electronic health data opening new possibilities in diagnostic radiology. In recent years, the performance of deep learning (DL) algorithms on various medical image tasks have continually improved. DL algorithms have been proposed as a tool to detect various forms of intracranial hemorrhage on non-contrast computed tomography (NCCT) of the head. In subtle, acute cases, the capacity for DL algorithm image interpretation support might improve the diagnostic yield of CT for detection of this time-critical condition, potentially expediting treatment where appropriate and improving patient outcomes. However, there are multiple challenges to DL algorithm implementation, such as the relative scarcity of labeled datasets, the difficulties in developing algorithms capable of volumetric medical image analysis, and the complex practicalities of deployment into clinical practice. This review examines the literature and the approaches taken in the development of DL algorithms for the detection of intracranial hemorrhage on NCCT head studies. Considerations in crafting such algorithms will be discussed, as well as challenges which must be overcome to ensure effective, dependable implementations as automated tools in a clinical setting.

Promises and pitfalls of deep neural networks in neuroimaging-based psychiatric research.

By promising more accurate diagnostics and individual treatment recommendations, deep neural networks and in particular convolutional neural networks have advanced to a powerful tool in medical imaging. Here, we first give an introduction into methodological key concepts and resulting methodological promises including representation and transfer learning, as well as modelling domain-specific priors. After reviewing recent applications within neuroimaging-based psychiatric research, such as the diagnosis of psychiatric diseases, delineation of disease subtypes, normative modeling, and the development of neuroimaging biomarkers, we discuss current challenges. This includes for example the difficulty of training models on small, heterogeneous and biased data sets, the lack of validity of clinical labels, algorithmic bias, and the influence of confounding variables.

Resilient asynchronous state estimation of Markov switching neural networks: A hierarchical structure approach.

This paper deals with the issue of resilient asynchronous state estimation of discrete-time Markov switching neural networks. Randomly occurring signal quantization and packet dropout are involved in the imperfect measured output. The asynchronous switching phenomena appear among Markov switching neural networks, quantizer modes and filter modes, which are modeled by a hierarchical structure approach. By resorting to the hierarchical structure approach and Lyapunov functional technique, sufficient conditions are achieved, and asynchronous resilient filters are derived such that filtering error dynamic is stochastically stable. Finally, two examples are included to verify the validity of the proposed method.

DeepTFactor: A deep learning-based tool for the prediction of transcription factors.

A transcription factor (TF) is a sequence-specific DNA-binding protein that modulates the transcription of a set of particular genes, and thus regulates gene expression in the cell. TFs have commonly been predicted by analyzing sequence homology with the DNA-binding domains of TFs already characterized. Thus, TFs that do not show homologies with the reported ones are difficult to predict. Here we report the development of a deep learning-based tool, DeepTFactor, that predicts whether a protein in question is a TF. DeepTFactor uses a convolutional neural network to extract features of a protein. It showed high performance in predicting TFs of both eukaryotic and prokaryotic origins, resulting in F1 scores of 0.8154 and 0.8000, respectively. Analysis of the gradients of prediction score with respect to input suggested that DeepTFactor detects DNA-binding domains and other latent features for TF prediction. DeepTFactor predicted 332 candidate TFs in Escherichia coli K-12 MG1655. Among them, 84 candidate TFs belong to the y-ome, which is a collection of genes that lack experimental evidence of function. We experimentally validated the results of DeepTFactor prediction by further characterizing genome-wide binding sites of three predicted TFs, YqhC, YiaU, and YahB. Furthermore, we made available the list of 4,674,808 TFs predicted from 73,873,012 protein sequences in 48,346 genomes. DeepTFactor will serve as a useful tool for predicting TFs, which is necessary for understanding the regulatory systems of organisms of interest. We provide DeepTFactor as a stand-alone program, available at https://bitbucket.org/kaistsystemsbiology/deeptfactor.