Structure-transferring edge-enhanced grid dehazing network.

The problem of image dehazing has received a great deal of attention in the computer vision community over the past two decades. Under haze conditions, due to the scattering of water vapor and dust particles in the air, the sharpness of the image is seriously reduced, making it difficult for many computer vision systems, such as those for object detection, object recognition, surveillance, driver assistance, etc. to do further process and operation. However, the previous dehazing methods usually have shortcomings such as poor brightness, color cast, removal of uncleanliness, halos, artifacts, and blurring. To address these problems, we propose a novel Structure-transferring Edge-enhanced Grid Dehazing Network (SEGDNet) in this study. An edge-preserving smoothing operator, a guided filter, is used to efficiently decompose the images into low-frequency image structure and high-frequency edges. The Low-frequency Grid Dehazing Subnetwork (LGDSn) is proposed to effectively preserve the low-frequency structure while dehazing. The High-frequency Edge Enhancement Subnetwork (HEESn) is also proposed to enhance the edges and details while removing the noise. The Low-and-High frequency Fusion Subnetwork (L&HFSn) is used to fuse the low-frequency and high-frequency results to obtain the final dehazed image. The experimental results on synthetic and real-world datasets demonstrate that our method outperforms the state-of-the-art methods in both qualitative and quantitative evaluations.

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.

Statins use and its impact in EGFR-TKIs resistance to prolong the survival of lung cancer patients: A Cancer registry cohort study in Taiwan.

Statins have been shown to be a beneficial treatment as chemotherapy and target therapy for lung cancer. This study aimed to investigate the effectiveness of statins in combination with epidermal growth factor receptor-tyrosine kinase inhibitor therapy for the resistance and mortality of lung cancer patients. A population-based cohort study was conducted using the Taiwan Cancer Registry database. From January 1, 2007, to December 31, 2012, in total 792 non-statins and 41 statins users who had undergone EGFR-TKIs treatment were included in this study. All patients were monitored until the event of death or when changed to another therapy. Kaplan-Meier estimators and Cox proportional hazards regression models were used to calculate overall survival. We found that the mortality was significantly lower in patients in the statins group compared with patients in the non-statins group (4-y cumulative mortality, 77.3%; 95% confidence interval (CI), 36.6%-81.4% vs. 85.5%; 95% CI, 78.5%-98%; P = .004). Statin use was associated with a reduced risk of death in patients the group who had tumor sizes <3 cm (hazard ratio [HR], 0.51, 95% CI, 0.29-0.89) and for patients in the group who had CCI scores <3 (HR, 0.6; 95% CI, 0.41-0.88; P = .009). In our study, statins were found to be associated with prolonged survival time in patients with lung cancer who were treated with EGFR-TKIs and played a synergistic anticancer role.

Ecotoxicological effect of ketamine: Evidence of acute, chronic and photolysis toxicity to Daphnia magna.

Ketamine has been increasingly used in medicine and has the potential for abuse or illicit use around the world. Ketamine cannot be removed by conventional wastewater treatment plants. Although ketamine and its metabolite norketamine have been detected to a significant degree in effluents and aquatic environments, their ecotoxicity effects in aquatic organisms remain undefined. In this study, we investigated the acute toxicity of ketamine and its metabolite, along with the chronic reproductive toxicity of ketamine (5-100µg/L) to Daphnia magna. Multiple environmental scenarios were also evaluated, including drug mixtures and sunlight irradiation toxicity. Ketamine and norketamine caused acute toxicity to D. magna, with half lethal concentration (LC50) values of 30.93 and 25.35mg/L, respectively, after 48h of exposure. Irradiated solutions of ketamine (20mg/L) significantly increased the mortality of D. magna; pre-irradiation durations up to 2h rapidly increased the death rate to 100%. A new photolysis byproduct (M.W. 241) of norketamine that accumulates during irradiation was identified for the first time. The relevant environmental concentration of ketamine produced significant reproductive toxicity effects in D. magna, as revealed by the reduction of the number of total live offspring by 33.6-49.8% (p < 0.05). The toxicity results indicate that the environmental hazardous risks of the relevant ketamine concentration cannot be ignored and warrant further examination.

Meta-analytic structural equation modeling for exploring workplace friendship, well-being, and organizational commitment.

BACKGROUND: Workplace friendships, which are informal relationships in the workplace, foster mutual trust, assistance, and emotional support and create a sense of belonging among coworkers. Employees with positive perceptions of their job responsibilities, psychological needs, and work environment experience happiness, satisfaction, and an overall sense of well-being. A heightened sense of well-being in the workplace positively affects employee satisfaction with job tasks and the work environment, reinforcing employees' sense of belonging and organizational commitment. OBJECTIVE: The study proposed and validated a theoretical model, with meta-analytic structural equation modeling employed to conduct quantitative, empirical research on workplace friendships, well-being, and organizational commitment in Taiwan and to analyze correlations between relevant variables. METHODS: The theoretical model was further validated through structural equation modeling, which yielded favorable goodness of fit in terms of various indicators. RESULTS: The mediating effect of well-being was verified through bootstrapping analysis. Within a 95% confidence interval, the total effect size of workplace friendships on organizational commitment was 0.549, which is the sum of its direct effect (0.255) and indirect effect (0.294). CONCLUSIONS: This finding indicates that well-being plays a mediating role in the association between workplace friendships and organizational commitment. The proposed theoretical model in this study is supported. Finally, the research results are discussed, and practical suggestions are provided.

Bronchoscopic management of bronchopleural fistula using free fat pad transplant with platelet-rich plasma: a case study.

BACKGROUND: A bronchopleural fistula (BPF) occurs when an abnormal connection forms between the bronchial tubes and pleural cavity, often due to surgery, infection, trauma, radiation, or chemotherapy. The outcomes of both surgical and bronchoscopic treatments frequently prove to be unsatisfactory. CASE PRESENTATION: Here, we report a case of successful bronchoscopic free fat pad transplantation combined with platelet-rich plasma, effectively addressing a post-lobectomy BPF. Contrast-enhanced chest tomography revealed pleural thickening with heterogeneous consolidations over the right upper and middle lobes, indicative of destructive lung damage and bronchiectasis. The patient underwent thoracoscopic bilobectomy of the lungs. During surgery, severe adhesions and calcification of the chest wall and lung parenchyma were observed. The entire hilar structure was calcified, presenting challenges for dissection, despite the assistance of energy devices. Bronchoscopic intervention was required, during which two abdominal subcutaneous fat pads were retrieved. CONCLUSION: This innovative approach offers promise in the management of BPF and signals potential advancements in enhancing treatment efficacy and patient recovery.

Solar-driven persulfate degradation of caffeine and cephradine in synthetic human urine.

Urine source separation, as an innovative concept for the reuse of microlevel nutrients in human urine, has drawn increasing attention recently. Consequently, removing coexisting pharmaceuticals in urine is necessary for further reuse. This study is the first to apply the solar-driven persulfate process (Solar/PS) to the investigation of cephradine (CFD) and caffeine (CAF) degradation in synthetic human urine. The results showed that significantly more degradation of CFD and CAF occurs with the Solar/PS process than with persulfate oxidation and direct sunlight photolysis, respectively. The generated reactive species ·OH, SO4·-, O2·- and 1O2 were identified in the Solar/PS process. While SO4·- played a dominant role at pH 6, it played a minor role at pH 9 due to the lower amount generated under alkaline conditions. The presence of chloride and ammonia negatively impacted the photodegradation of both compounds. In contrast, bicarbonate exhibited no effect on CAF but enhanced CFD degradation owing to its amino-acid-like structure, which has a higher reactivity toward CO3·-. Although total organic carbon (TOC) was partially mineralized after 6 h of operation, no Microtox® toxicity was observed.

The application of cardiac magnetic resonance imaging (CMR) in myocarditis after COVID-19 vaccines : Case series from single medical center in Taiwan.

During the coronavirus disease 2019 (COVID-19) pandemic, reports of vaccine-induced myocarditis, particularly messenger ribonucleic acid (mRNA)-based myocarditis, were widely spread. This case series describes various cases of COVID-19 vaccine-induced myocarditis confirmed by cardiac magnetic resonance imaging (MRI), including those who were administered rare protein-based vaccines. Eleven patients comprising eight males and three females with suspected myocarditis underwent cardiac MRI at Taichung Veterans General Hospital between October 2021 and May 2022. The median age of the patients was 33.5 years old (range: 22-57 years). The onset of myocarditis was mainly observed following mRNA vaccine inoculation. One patient received the MVC-COV1901 vaccine, a unique protein-based COVID-19 vaccine in Taiwan, and met the 2018 Lake Louise Criteria for the diagnosis of myocarditis, confirmed by cardiac MRI. Most patients reported chest discomfort after receiving various vaccine types. Among four patients with reduced left ventricular ejection fraction (LVEF), two showed LVEF restoration during the follow-up period, and the other two were lost to follow-up. Cardiac MRI characterizes myocardial features such as edema, inflammation, and fibrosis, and has been proven to diagnose myocarditis accurately with a sensitivity of 87.5% and a specificity of 96.2% according to the 2018 Lake Louise criteria. This diagnosis was achieved without invasive procedures such as endomyocardial biopsy or coronary angiography.

Evaluation of spatial and temporal changes in illicit drug use in the Taipei metropolitan area via wastewater-based epidemiology.

Taiwan, identified as pivotal in the Asian drug trafficking chain, has been experiencing a surge in illicit drug-related issues. Wastewater-based epidemiology (WBE) has emerged as a promising approach for comprehensive evaluation of actual illicit drug usage. This study presents the first WBE investigation of illicit drug consumption in Taiwan based on the analysis of wastewater from four wastewater treatment plants (WWTPs) in the Taipei metropolitan area. Additionally, it demonstrates a high correlation between the amounts of illicit drugs seized and influent concentrations over an extended period of time. The reliability of solid-phase extraction and analysis via high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) was validated for 16 illicit drugs (methamphetamine, ketamine, cocaine, codeine, methadone, morphine, meperidine, fentanyl, sufentanil, para-methoxyamphetamine (PMA), para-methoxymethamphetamine (PMMA), 3,4-methylenedioxymethamphetamine (MDMA), cathinone, methcathinone, mephedrone (MEPH), and 4-methylethcathinone (4-MEC)). Methamphetamine, ketamine, and 4-MEC were consistently detected in all wastewater samples, underscoring their prevalence in the Taipei metropolitan area. Biochemical oxygen demand (BOD) and ammonia nitrogen (ammonia N) were employed to reduce uncertainty in estimations of population size during back-calculation of illicit drug consumption. The results indicate that methamphetamine was the most consumed drug (175-740 mg day-1 1000 people-1), followed by ketamine (22-280 mg day-1 1000 people-1). In addition, urban-related WWTPs exhibited higher consumption of methamphetamine and ketamine than did the suburban-related WWTP, indicating distinct illicit drug usage patterns between suburban and urban regions. Moreover, an examination of temporal trends in wastewater from the Dihua WWTP revealed a persistent predominance of ketamine and methamphetamine, consistent with statistical data pertaining to seizure quantities and urine test results. The study provides encouraging insight into spatial and temporal variations in illicit drug usage in the Taipei metropolitan area, emphasizing the complementary role of WBE in understanding trends in illicit drug abuse.

Evolution of reactive species and their contribution to the removal of ketamine and amine-containing pharmaceuticals during the sunlight/chlorine process.

Sunlight-induced photoirradiation of chlorine (sunlight/chlorine) can be observed in outdoor swimming pools and open-channel disinfection units for wastewater treatment. In this study, the degradation of ketamine, an environmentally persistent pharmaceutical, under sunlight irradiation in the presence of a low concentration of chlorine (1 mg/L as Cl2) was investigated to elucidate the evolution of reactive species and their contribution to ketamine removal. •OH dominates the initial stage of sunlight/chlorine; however, after chlorine depletion, reactions still progress with an observed rate constant (kobs = 7.6 ± 0.50 × 10-3 min-1) an order of magnitude higher than photolysis alone (kobs = 2.9 ± 0.15 × 10-4 min-1). When chlorine is depleted, O3 becomes the major reactant that degrades ketamine. High O3 yields were found in both sunlight/HOCl (12.5 ± 0.5% at pH 5) and sunlight/ClO- (10 ± 1% at pH 10) systems. At sub-µM levels, O3 resulted in substantial removal of ketamine, and even faster rates were observed in the presence of sunlight. A kinetic model was also established, and evaluate time-dependent concentration levels during sunlight/chlorine. The model simulation showed that the cumulative O3 concentration could reach 0.91 µM, and O3 contributed 31% ketamine removal during the sunlight/chlorine process. Primary and secondary amine functional groups were demonstrated to be the reaction sites of O3; other pharmaceuticals, such as atenolol and metoprolol, underwent similar phenomena. In addition, the experimental and model results further indicated that sunlight/ClO2- or ClO2 also participates in the degradation of ketamine with a minor role; trace amounts (below nM level) of ClO2- and ClO2 were estimated by the simulation.

Reduction of Blood Loss by Means of the Cavitron Ultrasonic Surgical Aspirator for Thoracoscopic Salvage Anatomic Lung Resections.

In centrally located lung tumors, salvage pulmonary resections pose challenges due to adhesions between the pulmonary parenchyma, chest wall, and hilum. This study aimed to investigate the surgical outcomes associated with Cavitron Ultrasonic Surgical Aspirator (CUSA) usage in thoracoscopic salvage pulmonary resections. Patients with centrally located advanced-stage lung tumors who underwent salvage anatomic resections following systemic or radiotherapy were included. They were categorized into CUSA and non-CUSA groups, and perioperative parameters and surgical outcomes were analyzed. Results: The study included 7 patients in the CUSA group and 15 in the non-CUSA group. Despite a longer median surgical time in the CUSA group (3.8 h vs. 6.0 h, p = 0.021), there was a significant reduction in blood loss (100 mL vs. 250 mL, p = 0.014). Multivariate analyses revealed that the use of CUSA and radiotherapy had opposing effects on blood loss (ß: -296.7, 95% CI: -24.8 to -568.6, p = 0.034 and ß: 282.9, 95% CI: 19.7 to 546.3, p = 0.037, respectively). In conclusion, while using CUSA in the salvage anatomic resection of centrally located lung cancer may result in a longer surgical time, it is crucial in minimizing blood loss during the procedure.

Pulmonary Recruitment Prior to Intraoperative Multiple Pulmonary Ground-Glass Nodule Localization Increases the Localization Accuracy-A Retrospective Study.

The standard treatment for early-stage lung cancer is complete tumor excision by limited resection of the lung. Preoperative localization is used before video-assisted thoracoscopic surgery (VATS) to improve the accuracy of pulmonary nodule excision. However, lung atelectasis and hypoxia resulting from controlling apnea during the localization procedure may affect the localization accuracy. Pre-procedural pulmonary recruitment may improve the respiratory mechanics and oxygenation during localization. In this study, we investigated the potential benefits of pre-localization pulmonary recruitment prior to pulmonary ground-glass nodule localization in a hybrid operating room. We hypothesized that pre-localization pulmonary recruitment would increase the localization accuracy, improve oxygenation, and prevent the need for re-inflation during the localization procedure. We retrospectively enrolled patients with multiple pulmonary nodule localizations before surgical intervention in our hybrid operating room. We compared the localization accuracy between patients who had undergone pre-procedure pulmonary recruitment and patients who had not. Saturation, re-inflation rate, apnea time, procedure-related pneumothorax, and procedure time were also recorded as secondary outcomes. The patients who had undergone pre-procedure recruitment had better saturation, shorter procedure time, and higher localization accuracy. The pre-procedure pulmonary recruitment maneuver was effective in increasing regional lung ventilation, leading to improved oxygenation and localization accuracy.

Control of disinfection by-product formation using ozone-based advanced oxidation processes.

The effects of ozone dosage, water temperature and catalyst addition in an ozonation-fluidized bed reactor (O3/FBR) on treated water quality and on the control of chlorinated and ozonated disinfection by-products (DBPs) were investigated. A biofiltration column was used to evaluate its removal efficiency on biodegradable organic matter and to reduce DBP formation. The Dong-Gang River, polluted by agricultural and domestic wastewater in Pingtung, Taiwan, was used as the water source. The treated water quality in terms of dissolved organic carbon (DOC), biodegradable DOC, ultraviolet absorbance at 254 nm (UV254) and specific UV absorbance (SUVA) improved with increasing ozone and catalyst dosages. Catalytic ozonation was more effective than ozonation alone at reducing the formation of DBPs at a given dosage. Experimental results show that water temperature had little effect on the treated water quality with the O3/FBR system used in this study (p > 0.05). The combination of O3/FBR and the biofiltration process effectively decreased the amount ofDBP precursors. The concentration of total trihalomethanes (TTHMs) was less than the maximum contaminant level (MCL) requirement, which is 80 microg/L, for all treated waters and the concentration of five haloacetic acids (HAA5) fell below 60 microg/L with an ozone dosage higher than 2.5 mg/L.

Synthesis of shape-controlled monodisperse wurtzite CuIn(x)Ga(1-x)S2 semiconductor nanocrystals with tunable band gap.

Monodisperse wurtzite CuIn(x)Ga(1-x)S(2) nanocrystals have been synthesized over the entire composition range using a facile solution-based method. Depending on the chemical composition and synthesis conditions, the morphology of the nanocrystals can be controlled in the form of bullet-like, rod-like, and tadpole-like shapes. The band gap of the nanocrystals increases linearly with increasing Ga concentration, with band gap values for the end members being close to those observed in the bulk. Colloidal suspensions of the nanocrystals are attractive for use as inks for low-cost fabrication of thin film solar cells by spin or spray coating.

Colloidal synthesis of magnetic CuCr2S4 nanocrystals and nanoclusters.

Nanocrystals and nanoclusters of the room-temperature magnetic spinel CuCr(2)S(4) have been synthesized using a facile solution-based method. The synthesis involves hot injection of an excess of 1-dodecanethiol (1-DDT) into a boiling coordinating solvent containing CuCl(2) and CrCl(3)·6H(2)O. Using octadecylamine (ODA) as a solvent yields cube-shaped nanocrystals with an average size of 20 ± 2 nm, while with oleylamine (OLA), nanoclusters with an average size of 31 ± 2.5 nm are obtained. In both cases, powder X-ray diffraction patterns confirmed the formation of the pure spinel phase without any impurities. While the synthesized powders are superparamagnetic near room temperature, they exhibit ferromagnetic behavior at lower temperatures, with magnetization (M(S)) values of 30 emu/g (1.63 µ(B)/f.u.) and 33 emu/g (1.79 µ(B)/f.u.) for the ODA- and OLA-capped nanocrystals and nanoclusters, respectively, at 5 K.

Wireless remote weather monitoring system based on MEMS technologies.

This study proposes a wireless remote weather monitoring system based on Micro-Electro-Mechanical Systems (MEMS) and wireless sensor network (WSN) technologies comprising sensors for the measurement of temperature, humidity, pressure, wind speed and direction, integrated on a single chip. The sensing signals are transmitted between the Octopus II-A sensor nodes using WSN technology, following amplification and analog/digital conversion (ADC). Experimental results show that the resistance of the micro temperature sensor increases linearly with input temperature, with an average TCR (temperature coefficient of resistance) value of 8.2 × 10(-4) (°C(-1)). The resistance of the pressure sensor also increases linearly with air pressure, with an average sensitivity value of 3.5 × 10(-2) (Ω/kPa). The sensitivity to humidity increases with ambient temperature due to the effect of temperature on the dielectric constant, which was determined to be 16.9, 21.4, 27.0, and 38.2 (pF/%RH) at 27 °C, 30 °C, 40 °C, and 50 °C, respectively. The velocity of airflow is obtained by summing the variations in resistor response as airflow passed over the sensors providing sensitivity of 4.2 × 10(-2), 9.2 × 10(-2), 9.7 × 10(-2) (Ω/ms(-1)) with power consumption by the heating resistor of 0.2, 0.3, and 0.5 W, respectively. The passage of air across the surface of the flow sensors prompts variations in temperature among each of the sensing resistors. Evaluating these variations in resistance caused by the temperature change enables the measurement of wind direction.

Enhanced sorption of the UV filter 4-methylbenzylidene camphor on aged PET microplastics from both experimental and theoretical perspectives.

In this study, the morphology and sorption behavior of polyethylene terephthalate (PET) microplastics during the aging process are investigated. To clarify the sorption mechanism of aged PET microplastics, the common sunblock 4-methylbenzylidene camphor (4-MBC) was chosen as the target contaminant, and UV irradiation was used for the laboratory aging simulation. The results show that oxygen-containing functional groups (carboxylic, carbonyl, ketone and hydroxyl groups) increase on the surface of aged PET microplastics. Based on density functional theory (DFT) simulations, the camphor part of 4-MBC acts as a hydrogen bond acceptor, whereas the carboxylic group on aged PET microplastics acts as a hydrogen bond donor. The formation of hydrogen bonding causes increased sorption of 4-MBC on aged PET microplastics. The sorption capacity increased from 5 to 11 µg g-1 for 50 ppb 4-MBC with 100 mg PET microplastics after a five-day aging process. Other environmental factors that affect sorption were also identified; a higher pH value and the presence of salinity reduced the amount of sorption. The sorption of virgin PET ranged from 8.0 to 3.4 µg g-1 and the sorption of aged PET ranged from 22 to 5 µg g-1 at pH 4 to 10. In the presence of salinity (10% seawater), the virgin PET sorption dropped to 2.1 µg g-1 while the aged PET sorption dropped to 4 µg g-1. A similar phenomenon was also observed in the sorption behavior under natural sunlight (the sorption of PET increased from 0.4 to 0.8 µg g-1 after 6 months of aging). The potential risk to ecosystems of aged PET microplastics under prolonged sunlight exposure in the natural environment could be greater than that predicted for virgin microplastics.

Incorporating Patient Preferences into a Decision-Making Model of Hand Trauma Reconstruction.

BACKGROUND: Few studies have addressed patient preferences in emergent surgical decision making. AIM OF THE STUDY: Analyzing patient preferences for hand trauma reconstruction to propose a decision-making model. METHODS: A conjoint analysis survey was developed with Sawtooth Software. Three common flaps-i.e., a cross-finger flap (CFF), a dorsal metacarpal artery perforator flap (DMAPF), and an arterialized venous flap (AVF)-were listed as treatment alternatives. Five attributes corresponding to these flaps were recovery time, total procedure, postoperative care methods, postoperative scar condition, and complication rate. Utility and importance scores were generated from the software, and preference characteristics were evaluated using cluster analysis. RESULTS: The survey was completed by 197 participants with hand trauma. Complication risk received the highest importance score (42.87%), followed by scar condition (21.55%). Cluster analysis classified the participants as "conservative," "practical," and "dual-concern". The dual-concern and conservative groups had more foreign laborers and highly educated participants, respectively, than the other groups. Most participants in the conservative and practical groups preferred DMAPF, whereas those in the dual-concern group favored CFF. Our proposed model consisted of shared decision-making and treatment recommendation pathways. CONCLUSION: Incorporating patient preferences into the decision-making model can strengthen patient-centered care. Further research on the applications of the proposed model is warranted.

Deep Learning Approach for the Development of a Novel Predictive Model for Prostate Cancer.

We developed a deep learning approach for accurate prediction of PCA patients one year earlier with minimal features from electronic health records. The area under the receiver operating curve for prediction of PCA was 0.94. Moreover, the sensitivity and specificity of CNN were 0.87 and 0.88, respectively.