Molecular characterization of a novel mycovirus from binucleate Rhizoctonia AG-A strain A46.

The complete genome sequence of a positive-sense single-stranded RNA (+ ssRNA) virus, Rhizoctonia beny-like virus 1 (RBLV1), isolated from binucleate Rhizoctonia AG-A strain A46, was determined. The RBLV1 genome is 10,280 nt in length and contains a short stretch of adenines at the 3' terminus. It contains a single open reading frame (ORF) encoding a 376.30-kDa protein with viral helicase and RNA-dependent RNA polymerase (RdRp) motifs. The encoded protein exhibited the highest sequence similarity to Rhizoctonia cerealis beny-like virus 0928-1 (RcBeLV 0928-1, 45.25%), with a sequence coverage of 63%. Phylogenetic analysis based on ORF protein sequences revealed that RBLV1 is a novel unclassified mycovirus.

Biological characterization of emerging fungal pathogen Colletotrichum associated with mango (Mangifera indica L.) post-harvest anthracnose from Vietnam.

BACKGROUND: Post-harvest anthracnose (PHA) of mango is a devastating disease, which results in huge loss to mango producers and importers. Various species of PHA, diverse pathogenicity, and different resistance towards fungicides make it essential to evaluate the pathogen taxonomic status and biological characterization. METHODS AND RESULTS: Two strains DM-1 and DM-2 isolated from the fruit of DaQing mango from Vietnam were identified as Colletotrichum fructicola and C. asianum respectively, based on the morphological features, along with the phylogenetic tree of ITS and ApMat combined sequences. The growth status of different Colletotrichum strains under different conditions was analyzed to reveal the biological characteristics. The optimum growth temperature of DM-1 and DM-2 was 28 °C and mycelia grew rapidly in the dark. Both strains could grow in media with pH 4-11, while the optimum pH value was 6. Maltose and soluble starch were the most suitable carbon source for DM-1 and DM-2 respectively, and the peptone was the most suitable nitrogen source for both strains. The lethal temperatures were recorded as 55 °C 5 min for DM-1, and 50 °C 10 min for DM-2. CONCLUSIONS: To the best of our knowledge, it is the first study reporting the identification of the pathogens: C. fructicola and C. asianum responsible for postharvest fruit anthracnose of mango in Vietnam.

Altered hepatic lipid droplet morphology and lipid metabolism in fasted Plin2-null mice.

Perilipin 2 (Plin2) binds to the surface of hepatic lipid droplets (LDs) with expression levels that correlate with triacylglyceride (TAG) content. We investigated if Plin2 is important for hepatic LD storage in fasted or high-fat diet-induced obese Plin2+/+ and Plin2-/- mice. Plin2-/- mice had comparable body weights, metabolic phenotype, glucose tolerance, and circulating TAG and total cholesterol levels compared with Plin2+/+ mice, regardless of the dietary regime. Both fasted and high-fat fed Plin2-/- mice stored reduced levels of hepatic TAG compared with Plin2+/+ mice. Fasted Plin2-/- mice stored fewer but larger hepatic LDs compared with Plin2+/+ mice. Detailed hepatic lipid analysis showed substantial reductions in accumulated TAG species in fasted Plin2-/- mice compared with Plin2+/+ mice, whereas cholesteryl esters and phosphatidylcholines were increased. RNA-Seq revealed minor differences in hepatic gene expression between fed Plin2+/+ and Plin2-/- mice, in contrast to marked differences in gene expression between fasted Plin2+/+ and Plin2-/- mice. Our findings demonstrate that Plin2 is required to regulate hepatic LD size and storage of neutral lipid species in the fasted state, while its role in obesity-induced steatosis is less clear.

Fatty acids abrogate the growth-suppressive effects induced by inhibition of cholesterol flux in pancreatic cancer cells.

BACKGROUND: Despite therapeutic advances, the prognosis of pancreatic ductal adenocarcinoma (PDAC) remains extremely poor. Metabolic reprogramming is increasingly recognized as a key contributor to tumor progression and therapy resistance in PDAC. One of the main metabolic changes essential for tumor growth is altered cholesterol flux. Targeting cholesterol flux appears an attractive therapeutic approach, however, the complex regulation of cholesterol balance in PDAC cells remains poorly understood. METHODS: The lipid content in human pancreatic duct epithelial (HPDE) cells and human PDAC cell lines (BxPC-3, MIA PaCa-2, and PANC-1) was determined. Cells exposed to eight different inhibitors targeting different regulators of lipid flux, in the presence or absence of oleic acid (OA) stimulation were assessed for changes in viability, proliferation, migration, and invasion. Intracellular content and distribution of cholesterol was assessed. Lastly, proteome profiling of PANC-1 exposed to the sterol O-acyltransferase 1 (SOAT1) inhibitor avasimibe, in presence or absence of OA, was performed. RESULTS: PDAC cells contain more free cholesterol but less cholesteryl esters and lipid droplets than HPDE cells. Exposure to different lipid flux inhibitors increased cell death and suppressed proliferation, with different efficiency in the tested PDAC cell lines. Avasimibe had the strongest ability to suppress proliferation across the three PDAC cell lines. All inhibitors showing cell suppressive effect disturbed intracellular cholesterol flux and increased cholesterol aggregation. OA improved overall cholesterol balance, reduced free cholesterol aggregation, and reversed cell death induced by the inhibitors. Treatment with avasimibe changed the cellular proteome substantially, mainly for proteins related to biosynthesis and metabolism of lipids and fatty acids, apoptosis, and cell adhesion. Most of these changes were restored by OA. CONCLUSIONS: The study reveals that disturbing the cholesterol flux by inhibiting the actions of its key regulators can yield growth suppressive effects on PDAC cells. The presence of fatty acids restores intracellular cholesterol balance and abrogates the alternations induced by cholesterol flux inhibitors. Taken together, targeting cholesterol flux might be an attractive strategy to develop new therapeutics against PDAC. However, the impact of fatty acids in the tumor microenvironment must be taken into consideration.

Long-Term Efficacy and Safety of Recombinant Human Growth Hormone in Children Born Small for Gestational Age.

There is a lack of long-term data on the benefit of growth hormone (GH) treatment in Chinese children born small for gestational age (SGA). This study was conducted to assess the long-term efficacy and safety of GH treatment in children born SGA. One hundred and twenty prepubertal SGA children who did not achieve catch-up growth with height remained less than -2 standard deviations (SD) below gender-specific height were enrolled in this two-year, randomized, dose-comparative study followed by an extension study of up to 10 years. Daily subcutaneous injections of 0.23 mg/kg/week [low-dose (LD) group] or 0.46 mg/kg/week [high-dose (HD) group] somatropin were given for 104 weeks. Dosing in the extension study was≤0.46 mg/kg/week. The main outcome measures were change in height SD score (ΔHT-SDS), height velocity, insulin-like growth factor (IGF)-1, and IGF-1/IGF binding protein-3 (IGFBP-3) molar ratio. ΔHT-SDS at week 104 was 0.91±0.53 and 1.52±0.64 in the LD and HD groups (intergroup p<0.0001), respectively, and continued in an upward trend throughout the extension study, remaining above+2 for those who received treatment for a total of 7 years or more. At week 104, significant improvements were observed in height velocity, IGF-1 SDS, and IGF-1/IGFBP-3 molar ratio. Adult HT-SDS was -0.81±1.68 for boys and -0.82±1.05 for girls (p=0.9837). Glucose metabolism and thyroid function were within the normal reference range throughout treatment. Long-term recombinant human GH treatment was tolerable and effective at improving height in children born SGA.

Can Catenated Nitrogen Compounds with Amine-like Structures Become Candidates for High-Energy-Density Compounds?

The worthwhile idea of whether amine-like catenated nitrogen compounds are stable enough to be used as high-energy materials was proposed and answered. Abstracting the NH3 structure into NR3 (R is the substituent) yields a new class of amine-like catenated nitrogen compounds. Most of the azole ring structures have a high nitrogen content and stability. Inspired by this idea, a series of new amine-like catenated nitrogen compounds (A1 to H5) were designed, and their basic energetic properties were calculated. The results showed that (1) amine-like molecular structures are often characterized by low density; however, the density of these compounds increases as the number of nitrogens in the azole ring increases; (2) these catenated nitrogen compounds generally have extremely high enthalpies of formation (882.91-2652.03 kJ/mol), and the detonation velocity of some compounds exceeds 9254.00 m/s; (3) the detonation performance of amine-like catenated nitrogen compounds designed based on imidazole and pyrazole rings is poor due to their low nitrogen content; and (4) the bond dissociation enthalpy of trigger bonds of most compounds is higher than 84 kJ/mol, indicating that these compounds have a certain thermodynamic stability. In summary, amine-like catenated nitrogen compounds have the potential to become energetic compounds with excellent detonation properties and should be considered to be synthesized by experimental chemists.

Thermal stability of azole-rich energetic compounds: their structure, density, enthalpy of formation and energetic properties.

Energetic compounds, as a type of special material, are widely used in the fields of national defense, aerospace and exploration. Their research and production have received growing attention. Thermal stability is a crucial factor for the safety of energetic materials. Azole-rich energetic compounds have emerged as a research hotspot in recent years owing to their excellent properties. Due to the aromaticity of unsaturated azoles, many azole-rich energetic compounds have significant thermal stability, which is one of the properties that researchers focus on. This review presents a comprehensive summary of the physicochemical and energetic properties of various energetic materials, highlighting the relationship between thermal stability and the structural, physicochemical, and energetic properties of azole-rich energetic compounds. To improve the thermal stability of compounds, five aspects can be considered, including functional group modification, bridging, preparation of energetic salts, energetic metal-organic frameworks (EMOFs) and co-crystals. It was demonstrated that increasing the strength and number of hydrogen bonds of azoles and expanding the π-π stacking area are the key factors to improve thermal stability, which provides a valuable way to develop energetic materials with higher energy and thermal stability.

Regulation of stability and density of energetic materials via isomerism.

Selective regulation of stability and density via isomerism is a promising strategy for developing energetic materials. In this work, we selectively introduced dinitromethyl groups at different positions of 4-nitro-1,2,3-triazole. The regional heterogeneity endows a high crystal density by virtue of the dense packing; on the other hand, it changes the charge distribution in the molecule, and reinforces the hydrogen bonding interactions, all of which stabilize the material. The resulting compounds exhibit excellent detonation properties and impact sensitivity that are comparable to those of HMX (Dv = 9250 m s-1 and IS = 10 J).

Phylogenetic analyses of antimicrobial resistant Corynebacterium striatum strains isolated from a nosocomial outbreak in a tertiary hospital in China.

Corynebacterium striatum is an emerging, multidrug-resistant pathogen that frequently causes nosocomial infections worldwide. This study aimed to investigate phylogenetic relationship and presence of genes responsible for antimicrobial resistance among C. striatum strains associated with an outbreak at the Shanxi Bethune Hospital, China, in 2021. Fecal samples were collected from 65 patients with C. striatum infection at Shanxi Bethune Hospital between February 12, 2021 and April 12, 2021. C. striatum isolates were identified by 16S rRNA and rpoB gene sequencing. E-test strips were used to examine the antimicrobial susceptibility of the isolates. Whole-genome sequencing and bioinformatics analysis were employed to assess the genomic features and identify antimicrobial resistance genes of the isolates. Crystal violet staining was conducted to determine the ability of biofilm formation of each isolate. A total of 64 C. striatum isolates were identified and categorized into 4 clades based on single nucleotide polymorphisms. All isolates were resistant to penicillin, meropenem, ceftriaxone, and ciprofloxacin but susceptible to vancomycin and linezolid. Most isolates were also resistant to tetracycline, clindamycin, and erythromycin, with susceptibility rates of 10.77, 4.62, and 7.69%, respectively. Genomic analysis revealed 14 antimicrobial resistance genes in the isolates, including tetW, ermX, and sul1. Crystal violet staining showed that all isolates formed biofilms on the abiotic surface. Four clades of multidrug-resistant C. striatum spread in our hospitals possibly due to the acquisition of antimicrobial resistance genes.

Machine Learning Approaches for Predicting Psoriatic Arthritis Risk Using Electronic Medical Records: Population-Based Study.

BACKGROUND: Psoriasis (PsO) is a chronic, systemic, immune-mediated disease with multiorgan involvement. Psoriatic arthritis (PsA) is an inflammatory arthritis that is present in 6%-42% of patients with PsO. Approximately 15% of patients with PsO have undiagnosed PsA. Predicting patients with a risk of PsA is crucial for providing them with early examination and treatment that can prevent irreversible disease progression and function loss. OBJECTIVE: The aim of this study was to develop and validate a prediction model for PsA based on chronological large-scale and multidimensional electronic medical records using a machine learning algorithm. METHODS: This case-control study used Taiwan's National Health Insurance Research Database from January 1, 1999, to December 31, 2013. The original data set was split into training and holdout data sets in an 80:20 ratio. A convolutional neural network was used to develop a prediction model. This model used 2.5-year diagnostic and medical records (inpatient and outpatient) with temporal-sequential information to predict the risk of PsA for a given patient within the next 6 months. The model was developed and cross-validated using the training data and was tested using the holdout data. An occlusion sensitivity analysis was performed to identify the important features of the model. RESULTS: The prediction model included a total of 443 patients with PsA with earlier diagnosis of PsO and 1772 patients with PsO without PsA for the control group. The 6-month PsA risk prediction model that uses sequential diagnostic and drug prescription information as a temporal phenomic map yielded an area under the receiver operating characteristic curve of 0.70 (95% CI 0.559-0.833), a mean sensitivity of 0.80 (SD 0.11), a mean specificity of 0.60 (SD 0.04), and a mean negative predictive value of 0.93 (SD 0.04). CONCLUSIONS: The findings of this study suggest that the risk prediction model can identify patients with PsO at a high risk of PsA. This model may help health care professionals to prioritize treatment for target high-risk populations and prevent irreversible disease progression and functional loss.

Insensitive High-Energy Density Materials Based on Azazole-Rich Rings: 1,2,4-Triazole N-Oxide Derivatives Containing Isomerized Nitro and Amino Groups.

It is an arduous and meaningful challenge to design and develop new energetic materials with lower sensitivity and higher energy. How to skillfully combine the characteristics of low sensitivity and high energy is the key problem in designing new insensitive high-energy materials. Taking a triazole ring as a framework, a strategy of N-oxide derivatives containing isomerized nitro and amino groups was proposed to answer this question. Based on this strategy, some 1,2,4-triazole N-oxide derivatives (NATNOs) were designed and explored. The electronic structure calculation showed that the stable existence of these triazole derivatives was due to the intramolecular hydrogen bond and other interactions. The impact sensitivity and the dissociation enthalpy of trigger bonds directly indicated that some compounds could exist stably. The crystal densities of all NATNOs were larger than 1.80 g/cm3, which met the requirement of high-energetic materials for crystal density. Some NATNOs (9748 m/s for NATNO, 9841 m/s for NATNO-1, 9818 m/s for NATNO-2, 9906 m/s for NATNO-3, and 9592 m/s for NATNO-4) were potential high detonation velocity energy materials. These study results not only indicate that the NATNOs have relatively stable properties and excellent detonation properties but also prove that the strategy of nitro amino position isomerization coupled with N-oxide is an effective means to develop new energetic materials.

The Design, Synthesis and Application of Nitrogen Heteropolycyclic Compounds with UV Resistance Properties.

Exposure to ultraviolet (UV) light is known to cause skin aging, skin damage, cancer, and eye diseases, as well as polymer material aging. Therefore, significant attention has been devoted to the research and development of UV absorbers. Considering the robust hydrogen bonding and conjugated structure present in nitrogen-containing polycyclic compounds, these compounds have been selected as potential candidates for exploring ultraviolet absorption properties. After structural optimization and the simulation of ultraviolet absorption spectra, four tris-[1,2,4]-triazolo-[1,3,5]-triazine (TTTs) derivatives, namely TTTB, TTTD, TTTJ, and TTTL, were selected as the preferred compounds and synthesized. The structure of the compound was determined using various analytical techniques, including FTIR, 1HNMR, 13CNMR, HRMS, and XRD. Subsequently, composite films of polyvinyl chloride (PVC) and TTTs were produced using a simple solvent casting technique. The PVC films were subjected to UV age testing by exposing them to an ultraviolet aging chamber. The age-resistant performance of the fabricated films was evaluated using an ultraviolet spectrophotometer and Fourier infrared spectrum instrument. The findings suggest that TTTs exhibit a noteworthy capacity for absorbing ultraviolet radiation. The TTTL compound exhibits a superior UV absorption performance compared to commercially available UV absorbers such as UV-0 and UV-327 in the market.

Voice-based control system for smart hospital wards: a pilot study of patient acceptance.

BACKGROUND: The smart hospital's concept of using the Internet of Things (IoT) to reduce human resources demand has become more popular in the aging society. OBJECTIVE: To implement the voice smart care (VSC) system in hospital wards and explore patient acceptance via the Technology Acceptance Model (TAM). METHODS: A structured questionnaire based on TAM was developed and validated as a research tool. Only the patients hospitalized in the VSC wards and who used it for more than two days were invited to fill the questionnaire. Statistical variables were analyzed using SPSS version 24.0. A total of 30 valid questionnaires were finally obtained after excluding two incomplete questionnaires. Cronbach's α values for all study constructs were above 0.84. RESULT: We observed that perceived ease of use on perceived usefulness, perceived usefulness on user satisfaction and attitude toward using, and attitude toward using on behavioral intention to use had statistical significance (p < .01), respectively. CONCLUSION: We have successfully developed the VSC system in a Taiwanese academic medical center. Our study indicated that perceived usefulness was a crucial factor, which means the system function should precisely meet the patients' demands. Additionally, a clever system design is important since perceived ease of use positively affects perceived usefulness. The insight generated from this study could be beneficial to hospitals when implementing similar systems to their wards.

How Can Research on Artificial Empathy Be Enhanced by Applying Deepfakes?

We propose the idea of using an open data set of doctor-patient interactions to develop artificial empathy based on facial emotion recognition. Facial emotion recognition allows a doctor to analyze patients' emotions, so that they can reach out to their patients through empathic care. However, face recognition data sets are often difficult to acquire; many researchers struggle with small samples of face recognition data sets. Further, sharing medical images or videos has not been possible, as this approach may violate patient privacy. The use of deepfake technology is a promising approach to deidentifying video recordings of patients' clinical encounters. Such technology can revolutionize the implementation of facial emotion recognition by replacing a patient's face in an image or video with an unrecognizable face-one with a facial expression that is similar to that of the original. This technology will further enhance the potential use of artificial empathy in helping doctors provide empathic care to achieve good doctor-patient therapeutic relationships, and this may result in better patient satisfaction and adherence to treatment.

Pilot Report for Intracranial Hemorrhage Detection with Deep Learning Implanted Head Computed Tomography Images at Emergency Department.

Hemorrhagic stroke is a serious clinical condition that requires timely diagnosis. An artificial intelligence algorithm system called DeepCT can identify hemorrhagic lesions rapidly from non-contrast head computed tomography (NCCT) images and has received regulatory clearance. A non-controlled retrospective pilot clinical trial was conducted. Patients who received NCCT at the emergency department (ED) of Kaohsiung Veteran General Hospital were collected. From 2020 January-1st to April-30th, the physicians read NCCT images without DeepCT. From 2020May-1st to August-31st, the physicians were assisted by DeepCT. The length of ED stays (LOS) for the patients was collected. 2,999 patients were included (188 and 2811 with and without ICH). For patients with a final diagnosis of ICH, implementing DeepCT significantly shortened their LOS (560.67 ± 604.93 min with DeepCT vs. 780.83 ± 710.27 min without DeepCT; p = 0.0232). For patients with a non-ICH diagnosis, the LOS did not significantly differ (705.90 ± 760.86 min with DeepCT vs. 679.45 ± 681.97 min without DeepCT; p = 0.3362). For patients with ICH, those assisted with DeepCT had a significantly shorter LOS than those without DeepCT. For patients with a non-ICH diagnosis, implementing DeepCT did not affect the LOS, because emergency physicians need same efforts to identify the underlying problem(s) with or without DeepCT. In summary, implementing DeepCT system in the ED will save costs, decrease LOS, and accelerate patient flow; most importantly, it will improve the quality of care and increase the confidence and shorten the response time of the physicians and radiologists.

Plin2 deletion increases cholesteryl ester lipid droplet content and disturbs cholesterol balance in adrenal cortex.

Cholesteryl esters (CEs) are the water-insoluble transport and storage form of cholesterol. Steroidogenic cells primarily store CEs in cytoplasmic lipid droplet (LD) organelles, as contrasted to the majority of mammalian cell types that predominantly store triacylglycerol (TAG) in LDs. The LD-binding Plin2 binds to both CE- and TAG-rich LDs, and although Plin2 is known to regulate degradation of TAG-rich LDs, its role for regulation of CE-rich LDs is unclear. To investigate the role of Plin2 in the regulation of CE-rich LDs, we performed histological and molecular characterization of adrenal glands from Plin2+/+ and Plin2-/- mice. Adrenal glands of Plin2-/- mice had significantly enlarged organ size, increased size and numbers of CE-rich LDs in cortical cells, elevated cellular unesterified cholesterol levels, and increased expression of macrophage markers and genes facilitating reverse cholesterol transport. Despite altered LD storage, mobilization of adrenal LDs and secretion of corticosterone induced by adrenocorticotropic hormone stimulation or starvation were similar in Plin2+/+ and Plin2-/- mice. Plin2-/- adrenals accumulated ceroid-like structures rich in multilamellar bodies in the adrenal cortex-medulla boundary, which increased with age, particularly in females. Finally, Plin2-/- mice displayed unexpectedly high levels of phosphatidylglycerols, which directly paralleled the accumulation of these ceroid-like structures. Our findings demonstrate an important role of Plin2 for regulation of CE-rich LDs and cellular cholesterol balance in the adrenal cortex.

Risk of cancer in long-term levothyroxine users: Retrospective population-based study.

Levothyroxine is a widely prescribed medication for the treatment of an underactive thyroid. The relationship between levothyroxine use and cancer risk is largely underdetermined. To investigate the magnitude of the possible association between levothyroxine use and cancer risk, this retrospective case-control study was conducted using Taiwan's Health and Welfare Data Science Center database. Cases were defined as all patients who were aged ≥20 years and had a first-time diagnosis for cancer at any site for the period between 2001 and 2011. Multivariable conditional logistic regression models were used to calculate an adjusted odds ratio (AOR) to reduce potential confounding factors. A total of 601 733 cases and 2 406 932 controls were included in the current study. Levothyroxine users showed a 50% higher risk of cancer at any site (AOR: 1.50, 95% CI: 1.46-1.54; P < .0001) compared with non-users. Significant increased risks were also observed for brain cancer (AOR: 1.90, 95% CI: 1.48-2.44; P < .0001), skin cancer (AOR: 1.42, 95% CI: 1.17-1.72; P < .0001), pancreatic cancer (AOR: 1.27, 95% CI: 1.01-1.60; P = .03), and female breast cancer (AOR: 1.24, 95% CI: 1.15-1.33; P < .0001). Our study results showed that levothyroxine use was significantly associated with an increased risk of cancer, particularly brain, skin, pancreatic, and female breast cancers. Levothyroxine remains a highly effective therapy for hypothyroidism; therefore, physicians should carefully consider levothyroxine therapy and monitor patients' condition to avoid negative outcomes. Additional studies are needed to confirm these findings and to evaluate the potential biological mechanisms.

NFAT5 Regulated by STUB1, Facilitates Malignant Cell Survival and p38 MAPK Activation by Upregulating AQP5 in Chronic Lymphocytic Leukemia.

Chronic lymphocytic leukemia (CLL) is a clonal proliferative disease of mature B lymphocytes. To further improve the prognosis of patients, it is necessary to further elucidate the pathogenesis of CLL and find more effective therapeutic targets. Nuclear Factor of Activated T cells 5 (NFAT5) is the major activated transcription factor (TF) upon osmotic pressure increase in mammalian cells, and it also regulates many target genes to affect various cellular functions. The effects of NFAT5 on tumor growth and metastasis have also been widely revealed. However, the effects of NFAT5 on the progression of CLL are still unclear. In this study, we found abnormally high expression of NFAT5 in human CLL patients. Additionally, NFAT5 depletion suppressed proliferation and stimulated apoptosis of CLL cells. Our data further confirmed NFAT5 regulated AQP5 expression and the phosphorylation of p38 MAPK. We also found that AQP5 overexpression reversed the inhibitory effect of NFAT5 depletion on cell proliferation in CLL cells. Furthermore, we revealed STUB1 directly bound to NFAT5 and promoted its degradation. Taken together, our results indicate the involvement of NFAT5 in CLL progression and suggest that NFAT5 could serve as a promising therapeutic target for CLL treatment.

Correction: Artificial Intelligence-Based Prediction of Lung Cancer Risk Using Nonimaging Electronic Medical Records: Deep Learning Approach.

[This corrects the article DOI: 10.2196/26256.].

Artificial Intelligence-Based Prediction of Lung Cancer Risk Using Nonimaging Electronic Medical Records: Deep Learning Approach.

BACKGROUND: Artificial intelligence approaches can integrate complex features and can be used to predict a patient's risk of developing lung cancer, thereby decreasing the need for unnecessary and expensive diagnostic interventions. OBJECTIVE: The aim of this study was to use electronic medical records to prescreen patients who are at risk of developing lung cancer. METHODS: We randomly selected 2 million participants from the Taiwan National Health Insurance Research Database who received care between 1999 and 2013. We built a predictive lung cancer screening model with neural networks that were trained and validated using pre-2012 data, and we tested the model prospectively on post-2012 data. An age- and gender-matched subgroup that was 10 times larger than the original lung cancer group was used to assess the predictive power of the electronic medical record. Discrimination (area under the receiver operating characteristic curve [AUC]) and calibration analyses were performed. RESULTS: The analysis included 11,617 patients with lung cancer and 1,423,154 control patients. The model achieved AUCs of 0.90 for the overall population and 0.87 in patients ≥55 years of age. The AUC in the matched subgroup was 0.82. The positive predictive value was highest (14.3%) among people aged ≥55 years with a pre-existing history of lung disease. CONCLUSIONS: Our model achieved excellent performance in predicting lung cancer within 1 year and has potential to be deployed for digital patient screening. Convolution neural networks facilitate the effective use of EMRs to identify individuals at high risk for developing lung cancer.