Macrophage states: there's a method in the madness.

Single-cell approaches have shone a spotlight on discrete context-specific tissue macrophage states, deconstructed to their most minute details. Machine-learning (ML) approaches have recently challenged that dogma by revealing a context-agnostic continuum of states shared across tissues. Both approaches agree that 'brake' and 'accelerator' macrophage subpopulations must be balanced to achieve homeostasis. Both approaches also highlight the importance of ensemble fluidity as subpopulations switch between wide ranges of accelerator and brake phenotypes to mount the most optimal wholistic response to any threat. A full comprehension of the rules that govern these brake and accelerator states is a promising avenue because it can help formulate precise macrophage re-education therapeutic strategies that might selectively boost or suppress disease-associated states and phenotypes across various tissues.

The 100 most influential papers in medical artificial intelligence; a bibliometric analysis.

Objective: To assess the current trends in the field of artificial intelligence in medicine by analysing 100 most cited original articles relevant to the field. METHODS: The bibliometric analysis was conducted in September 2022, and comprised literature search on Scopus database for original articles only. Google and Medical Subject Headings databases were used as resources to extract key words. In order to cover a broad range of articles, original studies comprising human as well as non-human subjects, studies without abstract and studies in languages other than English were part of the inclusion criteria. There was no specific time period applied to the search and no specific selection was done regarding the journals in the database. The screening was done using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines to extract the top 100 most cited articles in the field of artificial intelligence usage in medicine. Data was analysed using SPSS 23. RESULTS: Of the 11,571 studies identified, 100(0.86%) were analysed in detail. The studies were published between 1986 and 2021, with a median of 43 citations (IQR 53) per article. The journal 'Artificial Intelligence in Medicine' accounted for the highest number 9(9%)) of articles, and the United States was the country of origin for most of the articles 36(36%). Conclusion: The trends, development and shortcomings in field of artificial intelligence usage in medicine need to be understood to conduct an effective research in areas that still need attention, and to guide the authorities to direct their funding accordingly.

Theory of trust and acceptance of artificial intelligence technology (TrAAIT): An instrument to assess clinician trust and acceptance of artificial intelligence.

BACKGROUND: Artificial intelligence and machine learning (AI/ML) technologies like generative and ambient AI solutions are proliferating in real-world healthcare settings. Clinician trust affects adoption and impact of these systems. Organizations need a validated method to assess factors underlying trust and acceptance of AI for clinical workflows in order to improve adoption and the impact of AI. OBJECTIVE: Our study set out to develop and assess a novel clinician-centered model to measure and explain trust and adoption of AI technology. We hypothesized that clinicians' system-specific Trust in AI is the primary predictor of both Acceptance (i.e., willingness to adopt), and post-adoption Trusting Stance (i.e., general stance towards any AI system). We validated the new model at an urban comprehensive cancer center. We produced an easily implemented survey tool for measuring clinician trust and adoption of AI. METHODS: This survey-based, cross-sectional, psychometric study included a model development phase and validation phase. Measurement was done with five-point ascending unidirectional Likert scales. The development sample included N = 93 clinicians (physicians, advanced practice providers, nurses) that used an AI-based communication application. The validation sample included N = 73 clinicians that used a commercially available AI-powered speech-to-text application for note-writing in an electronic health record (EHR). Analytical procedures included exploratory factor analysis (EFA), confirmatory factor analysis (CFA), and partial least squares structural equation modeling (PLS-SEM). The Johnson-Neyman (JN) methodology was used to determine moderator effects. RESULTS: In the fully moderated causal model, clinician trust explained a large amount of variance in their acceptance of a specific AI application (56%) and their post-adoption general trusting stance towards AI in general (36%). Moderators included organizational assurances, length of time using the application, and clinician age. The final validated instrument has 20 items and takes 5 min to complete on average. CONCLUSIONS: We found that clinician acceptance of AI is determined by their degree of trust formed via information credibility, perceived application value, and reliability. The novel model, TrAAIT, explains factors underlying AI trustworthiness and acceptance for clinicians. With its easy-to-use instrument and Summative Score Dashboard, TrAAIT can help organizations implementing AI to identify and intercept barriers to clinician adoption in real-world settings.

A pilot study of deep learning-based CT volumetry for traumatic hemothorax.

PURPOSE: We employ nnU-Net, a state-of-the-art self-configuring deep learning-based semantic segmentation method for quantitative visualization of hemothorax (HTX) in trauma patients, and assess performance using a combination of overlap and volume-based metrics. The accuracy of hemothorax volumes for predicting a composite of hemorrhage-related outcomes - massive transfusion (MT) and in-hospital mortality (IHM) not related to traumatic brain injury - is assessed and compared to subjective expert consensus grading by an experienced chest and emergency radiologist. MATERIALS AND METHODS: The study included manually labeled admission chest CTs from 77 consecutive adult patients with non-negligible (≥ 50 mL) traumatic HTX between 2016 and 2018 from one trauma center. DL results of ensembled nnU-Net were determined from fivefold cross-validation and compared to individual 2D, 3D, and cascaded 3D nnU-Net results using the Dice similarity coefficient (DSC) and volume similarity index. Pearson's r, intraclass correlation coefficient (ICC), and mean bias were also determined for the best performing model. Manual and automated hemothorax volumes and subjective hemothorax volume grades were analyzed as predictors of MT and IHM using AUC comparison. Volume cut-offs yielding sensitivity or specificity ≥ 90% were determined from ROC analysis. RESULTS: Ensembled nnU-Net achieved a mean DSC of 0.75 (SD: ± 0.12), and mean volume similarity of 0.91 (SD: ± 0.10), Pearson r of 0.93, and ICC of 0.92. Mean overmeasurement bias was only 1.7 mL despite a range of manual HTX volumes from 35 to 1503 mL (median: 178 mL). AUC of automated volumes for the composite outcome was 0.74 (95%CI: 0.58-0.91), compared to 0.76 (95%CI: 0.58-0.93) for manual volumes, and 0.76 (95%CI: 0.62-0.90) for consensus expert grading (p = 0.93). Automated volume cut-offs of 77 mL and 334 mL predicted the outcome with 93% sensitivity and 90% specificity respectively. CONCLUSION: Automated HTX volumetry had high method validity, yielded interpretable visual results, and had similar performance for the hemorrhage-related outcomes assessed compared to manual volumes and expert consensus grading. The results suggest promising avenues for automated HTX volumetry in research and clinical care.

New Opportunities and Challenges for Forensic Medicine in the Era of Artificial Intelligence Technology.

ABSTRACT: Traditional forensic identification relies on forensic experts to manually extract information and provide identification opinions based on medicine, biology and other fields of knowledge combined with personal work experience, which is not only time-consuming and require great effort, but also affected by subjective factors that are difficult to overcome. In the era of big data, the booming development of artificial intelligence brings new ideas to forensic medicine. In recent years, forensic researchers at home and abroad have conducted many studies based on artificial intelligence technology, such as face recognition, age and gender identification, DNA analysis, postmortem interval estimation, injury and cause of death identification, showing the feasibility and advantages of using artificial intelligence technology to solve forensic identification problems. As a new means of technology that has adapted to the development of the times, artificial intelligence has brought new vitality to forensic medicine, but at the same time also some new challenges. How to deal with these challenges scientifically and form a new mode of 'artificial intelligence plus forensic medicine' with artificial intelligence and forensic medicine developing collaboratively is a new direction for the development of forensic medicine in the era of big data.

Enhancing radiology training with GPT-4: Pilot analysis of automated feedback in trainee preliminary reports.

RATIONALE AND OBJECTIVES: Radiology residents often receive limited feedback on preliminary reports issued during independent call. This study aimed to determine if Large Language Models (LLMs) can supplement traditional feedback by identifying missed diagnoses in radiology residents' preliminary reports. MATERIALS & METHODS: A randomly selected subset of 500 (250 train/250 validation) paired preliminary and final reports between 12/17/2022 and 5/22/2023 were extracted and de-identified from our institutional database. The prompts and report text were input into the GPT-4 language model via the GPT-4 API (gpt-4-0314 model version). Iterative prompt tuning was used on a subset of the training/validation sets to direct the model to identify important findings in the final report that were absent in preliminary reports. For testing, a subset of 10 reports with confirmed diagnostic errors were randomly selected. Fourteen residents with on-call experience assessed the LLM-generated discrepancies and completed a survey on their experience using a 5-point Likert scale. RESULTS: The model identified 24 unique missed diagnoses across 10 test reports with i% model prediction accuracy as rated by 14 residents. Five additional diagnoses were identified by users, resulting in a model sensitivity of 79.2 %. Post-evaluation surveys showed a mean satisfaction rating of 3.50 and perceived accuracy rating of 3.64 out of 5 for LLM-generated feedback. Most respondents (71.4 %) favored a combination of LLM-generated and traditional feedback. CONCLUSION: This pilot study on the use of LLM-generated feedback for radiology resident preliminary reports demonstrated notable accuracy in identifying missed diagnoses and was positively received, highlighting LLMs' potential role in supplementing conventional feedback methods.

AI and machine learning in medical imaging: key points from development to translation.

Innovation in medical imaging artificial intelligence (AI)/machine learning (ML) demands extensive data collection, algorithmic advancements, and rigorous performance assessments encompassing aspects such as generalizability, uncertainty, bias, fairness, trustworthiness, and interpretability. Achieving widespread integration of AI/ML algorithms into diverse clinical tasks will demand a steadfast commitment to overcoming issues in model design, development, and performance assessment. The complexities of AI/ML clinical translation present substantial challenges, requiring engagement with relevant stakeholders, assessment of cost-effectiveness for user and patient benefit, timely dissemination of information relevant to robust functioning throughout the AI/ML lifecycle, consideration of regulatory compliance, and feedback loops for real-world performance evidence. This commentary addresses several hurdles for the development and adoption of AI/ML technologies in medical imaging. Comprehensive attention to these underlying and often subtle factors is critical not only for tackling the challenges but also for exploring novel opportunities for the advancement of AI in radiology.

Using machine learning for chemical-free histological tissue staining.

Hematoxylin and eosin staining can be hazardous, expensive, and prone to error and variability. To circumvent these issues, artificial intelligence/machine learning models such as generative adversarial networks (GANs), are being used to 'virtually' stain unstained tissue images indistinguishable from chemically stained tissue. Frameworks such as deep convolutional GANs (DCGAN) and conditional GANs (CGANs) have successfully generated highly reproducible 'stained' images. However, their utility may be limited by requiring registered, paired images which can be difficult to obtain. To avoid these dataset requirements, we attempted to use an unsupervised CycleGAN pix2pix model(5,6) to turn unpaired, unstained bright-field images into pathologist-approved digitally 'stained' images. Using formalin-fixed-paraffin-embedded liver samples, 5µm section images (20x) were obtained before and after staining to create "stained" an "unstained" datasets. Model implementation was conducted using Ubuntu 20.04.4 LTS, 32 GB RAM, Intel Core i7-9750 CPU @2.6 GHz, Nvidia GeForce RTX 2070 Mobile, Python 3.7.11 and Tensorflow 2.9.1. The CycleGAN framework utilized a u-net-based generator and discriminator from pix2pix, a CGAN. The CycleGAN used a modified loss function, cycle consistent loss that assumed unpaired images, so loss was measured twice. To our knowledge, this is the first documented application of this architecture using unpaired bright-field images. Results and suggested improvements are discussed.

Current status and challenges of model-informed drug discovery and development in China.

In the past decade, biopharmaceutical research and development in China has been notably boosted by government policies, regulatory initiatives and increasing investments in life sciences. With regulatory agency acting as a strong driver, model-informed drug development (MIDD) is transitioning rapidly from an academic pursuit to a critical component of innovative drug discovery and development within the country. In this article, we provided a cross-sectional summary on the current status of MIDD implementations across early and late-stage drug development in China, illustrated by case examples. We also shared insights into regulatory policy development and decision-making. Various modeling and simulation approaches were presented across a range of applications. Furthermore, the challenges and opportunities of MIDD in China were discussed and compared with other regions where these practices have a more established history. Through this analysis, we highlighted the potential of MIDD to enhance drug development efficiency and effectiveness in China's evolving pharmaceutical landscape.

Remote Monitoring, AI, Machine Learning and Mobile Ultrasound Integration upon 5G Internet in the Prehospital Care to Support the Golden Hour Principle and Optimize Outcomes in Severe Trauma and Emergency Surgery.

AIM: Feasibility and reliability evaluation of 5G internet networks (5G IN) upon Artificial Intelligence (AI)/Machine Learning (ML), of telemonitoring and mobile ultrasound (m u/s) in an ambulance car (AC)- integrated in the pre-hospital setting (PS)- to support the Golden Hour Principle (GHP) and optimize outcomes in severe trauma (TRS). MATERIAL AND METHODS: (PS) organization and care upon (5G IN) high bandwidths (10 GB/s) mobile tele-communication (mTC) experimentation by using the experimental Cobot PROMETHEUS III, pn:100016 by simulation upon six severe trauma clinical cases by ten (N1=10) experts: Four professional rescuers (n1=4), three trauma surgeons (n2=3), a radiologist (n3=1) and two information technology specialists (n4=2) to evaluate feasibility, reliability and clinical usability for instant risk, prognosis and triage computation, decision support and treatment planning by (AI)/(ML) computations in (PS) of (TRS) as well as by performing (PS) (m u/s). RESULTS: A. Trauma severity scales instant computations by the Cobot PROMETHEUS III, pn 100016) ) based on AI and ML complex algorithms and Cloud Computing, telemonitoring and r showed very high feasibility and reliability upon (5GIN) under specific, technological, training and ergonomic prerequisites B. Measured be-directional (m u/s) images data sharing between (AC) and (ED/TC) showed very high feasibility and reliability upon (5G IN) under specific, technological and ergonomic conditions in (TRS). CONCLUSION: Integration of (PS) tele-monitoring with (AI)/(ML) and (PS) (m u/s) upon (5GIN) via the Cobot PROMETHEUS III, (pn 100016) in severe (TRS/ES), seems feasible and under specific prerequisites reliable to support the (GHP) and optimize outcomes in adult and pediatric (TRS/ES).

Review on Uric Acid Recognition by MOFs with a Future in Machine Learning.

Uric acid (UA) is produced from purine metabolism and serves as a prevalent biomarker for multiple diseases including cancer. Hyperuricemia or hypouricemia can cause multiple dysfunctions throughout the biological processes. Consequently, there is a pressing need for monitoring UA concentration in body fluid. While clinical methods are known, the availability of a point-of-care testing (PoCT) kit remains conspicuously absent. In the case of electrochemical recognition of UA, the oxidation potential of ascorbic acid closely aligns with that of UA and thus it hinders the detection process, which eventually may result in false positive signals. Several chemosensors are known in the field of supramolecular chemistry, and metal-organic frameworks (MOFs) are one of the best-performing contenders due to their robustness, stability, and versatile structures. In this review, we tried to unbox the up-to-date development of UA sensing by MOFs. We delve into the state of UA recognition by MOFs, exploring both electrochemical and fluorometric pathways and drawing comparisons with structurally similar probes like covalent organic frameworks (COFs) to understand/establish the advantages of MOFs specifically in UA sensing. In the absence of a PoCT kit, we have provided the conceptual outlook for designing a PoCT device termed a "Urimeter" via electrochemical operation. For the first time, we have proposed different methods of how UA sensing can be tied up with artificial intelligence and machine learning (AI-ML).

Electrocardiogram-based deep learning model to screen peripartum cardiomyopathy.

BACKGROUND: Peripartum cardiomyopathy, one of the most fatal conditions during delivery, results in heart failure secondary to left ventricular systolic dysfunction. Left ventricular dysfunction can result in abnormalities in electrocardiography. However, the usefulness of electrocardiography in the identification of peripartum cardiomyopathy in pregnant women remains unclear. OBJECTIVE: This study aimed to evaluate the effectiveness of a 12-lead electrocardiography-based artificial intelligence/machine learning-based software as a medical device for screening peripartum cardiomyopathy. STUDY DESIGN: This retrospective cohort study included pregnant women who underwent transthoracic echocardiography between a month before and 5 months after delivery and underwent 12-lead electrocardiography within 30 days of echocardiography between December 2011 and May 2022 at Seoul National University Hospital. The performance of 12-lead electrocardiography-based artificial intelligence/machine learning analysis (AiTiALVSD software; version 1.00.00, which was developed to screen for left ventricular systolic dysfunction in the general population) was evaluated for the identification of peripartum cardiomyopathy. In addition, the performance of another artificial intelligence/machine learning algorithm using only 1-lead electrocardiography to detect left ventricular systolic dysfunction was evaluated in identifying peripartum cardiomyopathy. The results were obtained under a 95% confidence interval and considered significant when P<.05. RESULTS: Among the 14,557 women who delivered during the study period, 204 (1.4%) underwent transthoracic echocardiography a month before and 5 months after delivery. Among them, 12 (5.8%) were diagnosed with peripartum cardiomyopathy. The results showed that AiTiALVSD for 12-lead electrocardiography was highly effective in detecting peripartum cardiomyopathy, with an area under the receiver operating characteristic of 0.979 (95% confidence interval, 0.953-1.000), an area under the precision-recall curve of 0.715 (95% confidence interval, 0.499-0.951), a sensitivity of 0.917 (95% confidence interval, 0.760-1.000), a specificity of 0.927 (95% confidence interval, 0.890-0.964), a positive predictive value of 0.440 (95% confidence interval, 0.245-0.635), and a negative predictive value of 0.994 (95% confidence interval, 0.983-1.000). In addition, a 1-lead (lead I) artificial intelligence/machine learning algorithm showed excellent performance; the area under the receiver operating characteristic, area under the precision-recall curve, sensitivity, specificity, positive predictive value, and negative predictive value were 0.944 (95% confidence interval, 0.895-0.993), 0.520 (95% confidence interval, 0.319-0.801), 0.833 (95% confidence interval, 0.622-1.000), 0.880 (95% confidence interval, 0.834-0.926), 0.303 (95% confidence interval, 0.146-0.460), and 0.988 (95% confidence interval, 0.972-1.000), respectively. CONCLUSION: The 12-lead electrocardiography-based artificial intelligence/machine learning-based software as a medical device (AiTiALVSD) and 1-lead algorithm are noninvasive and effective ways of identifying cardiomyopathies occurring during the peripartum period, and they could potentially be used as highly sensitive screening tools for peripartum cardiomyopathy.

Unlocking the Power of Benchmarking: Real-World-Time Data Analysis for Enhanced Sarcoma Patient Outcomes.

Benchmarking is crucial for healthcare providers to enhance quality and efficiency, notably for complex conditions like sarcomas. Multidisciplinary teams/sarcoma boards (MDT/SBs) are vital in sarcoma management, but differences in their processes can affect patient outcomes and treatment costs, despite adherence to international guidelines. To address this issue, this study aimed to compare two MDT/SBs and establish an interoperable digital platform, Sarconnector®, for real-time-world data assessment and automated analysis. The study included 983 patients, 46.0% of whom female, with a median age of 58 years, and 4.5% of patients presented with metastasis at diagnosis. Differences were observed in the number of first-time presentations, follow-up presentations, primary sarcomas, biopsies and chemotherapy indications between the two MDT/SB. The results highlight the importance of benchmarking and utilizing a harmonized data approach, such as the RWT approach provided by the Sarconnector®, to standardize and evaluate quality and cost metrics. By identifying areas of improvement and making data-driven decisions on the meta-level, healthcare providers can optimize resources and improve patient outcomes. In conclusion, benchmarking with the RWT harmonized data approach provided by the Sarconnector® can help healthcare providers improve the overall effectiveness of the healthcare system and achieve better outcomes for their patients in terms of both outcomes and costs.

COVID-19 lung disease shares driver AT2 cytopathic features with Idiopathic pulmonary fibrosis.

BACKGROUND: In the aftermath of Covid-19, some patients develop a fibrotic lung disease, i.e., post-COVID-19 lung disease (PCLD), for which we currently lack insights into pathogenesis, disease models, or treatment options. METHODS: Using an AI-guided approach, we analyzed > 1000 human lung transcriptomic datasets associated with various lung conditions using two viral pandemic signatures (ViP and sViP) and one covid lung-derived signature. Upon identifying similarities between COVID-19 and idiopathic pulmonary fibrosis (IPF), we subsequently dissected the basis for such similarity from molecular, cytopathic, and immunologic perspectives using a panel of IPF-specific gene signatures, alongside signatures of alveolar type II (AT2) cytopathies and of prognostic monocyte-driven processes that are known drivers of IPF. Transcriptome-derived findings were used to construct protein-protein interaction (PPI) network to identify the major triggers of AT2 dysfunction. Key findings were validated in hamster and human adult lung organoid (ALO) pre-clinical models of COVID-19 using immunohistochemistry and qPCR. FINDINGS: COVID-19 resembles IPF at a fundamental level; it recapitulates the gene expression patterns (ViP and IPF signatures), cytokine storm (IL15-centric), and the AT2 cytopathic changes, e.g., injury, DNA damage, arrest in a transient, damage-induced progenitor state, and senescence-associated secretory phenotype (SASP). These immunocytopathic features were induced in pre-clinical COVID models (ALO and hamster) and reversed with effective anti-CoV-2 therapeutics in hamsters. PPI-network analyses pinpointed ER stress as one of the shared early triggers of both diseases, and IHC studies validated the same in the lungs of deceased subjects with COVID-19 and SARS-CoV-2-challenged hamster lungs. Lungs from tg-mice, in which ER stress is induced specifically in the AT2 cells, faithfully recapitulate the host immune response and alveolar cytopathic changes that are induced by SARS-CoV-2. INTERPRETATION: Like IPF, COVID-19 may be driven by injury-induced ER stress that culminates into progenitor state arrest and SASP in AT2 cells. The ViP signatures in monocytes may be key determinants of prognosis. The insights, signatures, disease models identified here are likely to spur the development of therapies for patients with IPF and other fibrotic interstitial lung diseases. FUNDING: This work was supported by the National Institutes for Health grants R01- GM138385 and AI155696 and funding from the Tobacco-Related disease Research Program (R01RG3780).

Artificial image objects for classification of schizophrenia with GWAS-selected SNVs and convolutional neural network.

In this article, we propose a new approach to analyze large genomics data. We considered individual genetic variants as pixels in an image and transformed a collection of variants into an artificial image object (AIO), which could be classified as a regular image by CNN algorithms. Using schizophrenia as a case study, we demonstrate the principles and their applications with 3 datasets. With 4,096 SNVs, the CNN models achieved an accuracy of 0.678 ± 0.007 and an AUC of 0.738 ± 0.008 for the diagnosis phenotype. With 44,100 SNVs, the models achieved class-specific accuracies of 0.806 ± 0.032 and 0.820 ± 0.049, and AUCs of 0.930 ± 0.017 and 0.867 ± 0.040 for the bottom and top classes stratified by the patient's polygenic risk scores. These results suggest that, once transformed to images, large genomics data can be analyzed effectively with image classification algorithms.

AI-guided discovery of the invariant host response to viral pandemics.

BACKGROUND: Coronavirus Disease 2019 (Covid-19) continues to challenge the limits of our knowledge and our healthcare system. Here we sought to define the host immune response, a.k.a, the "cytokine storm" that has been implicated in fatal COVID-19 using an AI-based approach. METHOD: Over 45,000 transcriptomic datasets of viral pandemics were analyzed to extract a 166-gene signature using ACE2 as a 'seed' gene; ACE2 was rationalized because it encodes the receptor that facilitates the entry of SARS-CoV-2 (the virus that causes COVID-19) into host cells. An AI-based approach was used to explore the utility of the signature in navigating the uncharted territory of Covid-19, setting therapeutic goals, and finding therapeutic solutions. FINDINGS: The 166-gene signature was surprisingly conserved across all viral pandemics, including COVID-19, and a subset of 20-genes classified disease severity, inspiring the nomenclatures ViP and severe-ViP signatures, respectively. The ViP signatures pinpointed a paradoxical phenomenon wherein lung epithelial and myeloid cells mount an IL15 cytokine storm, and epithelial and NK cell senescence and apoptosis determine severity/fatality. Precise therapeutic goals could be formulated; these goals were met in high-dose SARS-CoV-2-challenged hamsters using either neutralizing antibodies that abrogate SARS-CoV-2•ACE2 engagement or a directly acting antiviral agent, EIDD-2801. IL15/IL15RA were elevated in the lungs of patients with fatal disease, and plasma levels of the cytokine prognosticated disease severity. INTERPRETATION: The ViP signatures provide a quantitative and qualitative framework for titrating the immune response in viral pandemics and may serve as a powerful unbiased tool to rapidly assess disease severity and vet candidate drugs. FUNDING: This work was supported by the National Institutes for Health (NIH) [grants CA151673 and GM138385 (to DS) and AI141630 (to P.G), DK107585-05S1 (SD) and AI155696 (to P.G, D.S and S.D), U19-AI142742 (to S. C, CCHI: Cooperative Centers for Human Immunology)]; Research Grants Program Office (RGPO) from the University of California Office of the President (UCOP) (R00RG2628 & R00RG2642 to P.G, D.S and S.D); the UC San Diego Sanford Stem Cell Clinical Center (to P.G, D.S and S.D); LJI Institutional Funds (to S.C); the VA San Diego Healthcare System Institutional funds (to L.C.A). GDK was supported through The American Association of Immunologists Intersect Fellowship Program for Computational Scientists and Immunologists. ONE SENTENCE SUMMARY: The host immune response in COVID-19.

Artificial intelligence in radiology: the ecosystem essential to improving patient care.

The rapid development of artificial intelligence (AI) has led to its widespread use in multiple industries, including healthcare. AI has the potential to be a transformative technology that will significantly impact patient care. Particularly, AI has a promising role in radiology, in which computers are indispensable and new technological advances are often sought out and adopted early in clinical practice. We present an overview of the basic definitions of common terms, the development of an AI ecosystem in imaging and its value in mitigating the challenges of implementation in clinical practice.

Artificial Intelligence for Understanding Imaging, Text, and Data in Gastroenterology.

Artificial intelligence (AI) could change the practice of gastroenterology through its ability to both acquire and analyze information with speed, reproducibility, and, potentially, insight that may exceed that of human medical specialists. AI is powered by computational methods that allow machines to replicate clinical pattern recognition used by gastroenterology specialists to interpret endoscopic or cross-sectional images; understand the meaning and intent of medical documents; and merge different types of data to infer a diagnosis, prognosis, or expected outcome. Ongoing research is studying the use of AI for automated interpretation of text from colonoscopy and clinical documents for improved quality and patient phenotyping as well as enhanced detection and descriptions of polyps and other endoscopic lesions, and for predicting the probability of future therapeutic response early in a treatment course. This article introduces emerging technologies of natural language processing, machine vision, and machine learning for data analytics, and describes current and future applications in gastroenterology.

Artificial Intelligence in Radiotherapy Treatment Planning: Present and Future.

Treatment planning is an essential step of the radiotherapy workflow. It has become more sophisticated over the past couple of decades with the help of computer science, enabling planners to design highly complex radiotherapy plans to minimize the normal tissue damage while persevering sufficient tumor control. As a result, treatment planning has become more labor intensive, requiring hours or even days of planner effort to optimize an individual patient case in a trial-and-error fashion. More recently, artificial intelligence has been utilized to automate and improve various aspects of medical science. For radiotherapy treatment planning, many algorithms have been developed to better support planners. These algorithms focus on automating the planning process and/or optimizing dosimetric trade-offs, and they have already made great impact on improving treatment planning efficiency and plan quality consistency. In this review, the smart planning tools in current clinical use are summarized in 3 main categories: automated rule implementation and reasoning, modeling of prior knowledge in clinical practice, and multicriteria optimization. Novel artificial intelligence-based treatment planning applications, such as deep learning-based algorithms and emerging research directions, are also reviewed. Finally, the challenges of artificial intelligence-based treatment planning are discussed for future works.