RESUMEN
BACKGROUND: The COVID-19 pandemic has limited daily activities and even contact between patients and primary care providers. This makes it more difficult to provide adequate primary care services, which include connecting patients to an appropriate medical specialist. A smartphone-compatible artificial intelligence (AI) chatbot that classifies patients' symptoms and recommends the appropriate medical specialty could provide a valuable solution. OBJECTIVE: In order to establish a contactless method of recommending the appropriate medical specialty, this study aimed to construct a deep learning-based natural language processing (NLP) pipeline and to develop an AI chatbot that can be used on a smartphone. METHODS: We collected 118,008 sentences containing information on symptoms with labels (medical specialty), conducted data cleansing, and finally constructed a pipeline of 51,134 sentences for this study. Several deep learning models, including 4 different long short-term memory (LSTM) models with or without attention and with or without a pretrained FastText embedding layer, as well as bidirectional encoder representations from transformers for NLP, were trained and validated using a randomly selected test data set. The performance of the models was evaluated on the basis of the precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC). An AI chatbot was also designed to make it easy for patients to use this specialty recommendation system. We used an open-source framework called "Alpha" to develop our AI chatbot. This takes the form of a web-based app with a frontend chat interface capable of conversing in text and a backend cloud-based server application to handle data collection, process the data with a deep learning model, and offer the medical specialty recommendation in a responsive web that is compatible with both desktops and smartphones. RESULTS: The bidirectional encoder representations from transformers model yielded the best performance, with an AUC of 0.964 and F1-score of 0.768, followed by LSTM model with embedding vectors, with an AUC of 0.965 and F1-score of 0.739. Considering the limitations of computing resources and the wide availability of smartphones, the LSTM model with embedding vectors trained on our data set was adopted for our AI chatbot service. We also deployed an Alpha version of the AI chatbot to be executed on both desktops and smartphones. CONCLUSIONS: With the increasing need for telemedicine during the current COVID-19 pandemic, an AI chatbot with a deep learning-based NLP model that can recommend a medical specialty to patients through their smartphones would be exceedingly useful. This chatbot allows patients to identify the proper medical specialist in a rapid and contactless manner, based on their symptoms, thus potentially supporting both patients and primary care providers.
Asunto(s)
COVID-19/epidemiología , Aprendizaje Profundo , Atención Primaria de Salud/métodos , Derivación y Consulta , Teléfono Inteligente , Telemedicina/métodos , Humanos , Pandemias , SARS-CoV-2/aislamiento & purificación , EspecializaciónRESUMEN
BACKGROUND: The aim of this study was to report a rare case of nasopalatine duct cyst with sebaceous differentiation. Further, a systematic search of the literature was performed to identify studies reporting patients with intraosseous jaw cysts with sebaceous differentiation. CASE PRESENTATION: A 55-year-old Korean man was referred to our hospital because of a cystic lesion of the anterior maxilla. Radiologic examination revealed a well-circumscribed radiolucent lesion in the anterior maxilla. Histology showed a respiratory columnar and cuboidal epithelium-lined cyst. Transition from the ciliated columnar epithelium to stratified squamous epithelium with sebaceous differentiation was observed. Based on these findings, the final diagnosis was nasopalatine duct cyst with sebaceous differentiation. A systematic search of the literature was performed to identify studies reporting patients with intraosseous jaw cysts with sebaceous differentiation. There were 24 cases of sebaceous differentiation in the epithelium of the cysts including 2 odontogenic keratocysts, 8 orthokeratinized odontogenic cysts, 8 dentigerous cysts, 1 radicular cyst, and 2 glandular odontogenic cysts. However, no case reports describing the occurrence of nasopalatine duct cysts with sebaceous differentiation have been reported. CONCLUSION: This first case report of nasopalatine duct cysts with sebaceous differentiation could provide insight into the diagnostic process of cystic lesions with sebaceous differentiation.
Asunto(s)
Quistes Maxilomandibulares , Quistes Odontogénicos , Tumores Odontogénicos , Quiste Radicular , Epitelio , Humanos , Masculino , Persona de Mediana Edad , Quistes Odontogénicos/diagnóstico por imagenRESUMEN
Biopolymer adsorption onto a membrane is a significant issue in the reliability of solid-state nanopore devices, since it degrades the device performance or promotes device failure. In this work, a poly(ethylene glycol) (PEG) layer was coated on a silicon nitride (SiNx) membrane by plasma-polymerized vapor deposition to inhibit biopolymer adsorption. From optical observations, the deposited PEG layer demonstrated increased hydrophilicity and anti-adsorption property compared to the SiNx surface. Electrical properties of the PEG/SiNx nanopore were characterized, showing Ohmic behavior and a 6.3 times higher flicker noise power due to the flexible conformation of PEG in water. Antifouling performance of each surface was analyzed by measuring the average time from voltage bias to the first adsorption during DNA translocation experiments, where the modified surface enabled two times prolonged device operation. The time to adsorption was dependent on the applied voltage, implying adsorption probability was dominated by the electrophoretic DNA approach to the nanopore. DNA translocation behaviors on each surface were identified from translocation signals, as the PEG layer promoted unfolded and fast movement of DNA through the nanopore. This work successfully analyzed the effect of the PEG layer on DNA adsorption and translocation in solid-state nanopore experiments.
RESUMEN
Nanopore structures were originally proposed for detection of biomolecule translocation through nanometer-scale pores that perforate membranes by transient changes in ionic current. In this study, these changes are utilized to detect corrosion of different metals in aqueous chlorine media. The corrosion behaviors of Cu, Al, Ti, and Cr were analyzed by monitoring the changes in ion current resulting from ion concentration variations in solutions due to corrosion of the metals. We observed that the Cu layer passivated by CuO x was severely corroded when a drastic change of ion current was induced, from 10 to 30 nS to the level of 104 nS, as soon as the bias voltage was applied. In the case of Al passivated by thin AlO x , the conductance increased from 10-30 to 166 ± 52 nS and became saturated. Highly localized pitting corrosion was observed on the periphery of the nanopore, where the electrical field was most concentrated. Finally, we observed that Ti and Cr passivated by oxide showed long-term stability without corrosion in 1 M KCl in the pH range of 4-11.
RESUMEN
Memristors, or memristive devices, have attracted tremendous interest in neuromorphic hardware implementation. However, the high electric-field dependence in conventional filamentary memristors results in either digital-like conductance updates or gradual switching only in a limited dynamic range. Here, we address the switching parameter, the reduction probability of Ag cations in the switching medium, and ultimately demonstrate a cluster-type analogue memristor. Ti nanoclusters are embedded into densified amorphous Si for the following reasons: low standard reduction potential, thermodynamic miscibility with Si, and alloy formation with Ag. These Ti clusters effectively induce the electrochemical reduction activity of Ag cations and allow linear potentiation/depression in tandem with a large conductance range (~244) and long data retention (~99% at 1 hour). Moreover, according to the reduction potentials of incorporated metals (Pt, Ta, W, and Ti), the extent of linearity improvement is selectively tuneable. Image processing simulation proves that the Ti4.8%:a-Si device can fully function with high accuracy as an ideal synaptic model.
Asunto(s)
Ingeniería , Metales , Aleaciones , Simulación por Computador , Oxidación-ReducciónRESUMEN
A method to directly collect negatively charged nucleic acids, such as DNA and RNA, in the biosamples simply by applying an electric field in between the sample and collection buffer separated by the nanofilter membrane is proposed. The nanofilter membrane was made of low-stress silicon nitride with a thickness of 100 nm, and multiple pores were perforated in a highly arranged pattern using nanoimprint technology with a pore size of 200 nm and a pore density of 7.22 × 108/cm2. The electrophoretic transport of hsa-mir-93-5p across the membrane was confirmed in pure microRNA (miRNA) mimic solution using quantitative reverse transcription-polymerase chain reactions (qRT-PCR). Consistency of the collected miRNA quantity, stability of the system during the experiment, and yield and purity of the prepared sample were discussed in detail to validate the effectiveness of the electrical protocol. Finally, in order to check the applicability of this method to clinical samples, liquid biopsy process was demonstrated by evaluating the miRNA levels in sera of hepatocellular carcinoma patients and healthy controls. This efficient system proposed a simple, physical idea in preparation of nucleic acid from biosamples, and demonstrated its compatibility to biological downstream applications such as qRT-PCR as the conventional nucleic acid extraction protocols.
RESUMEN
Rapid and efficient nucleic acid (NA) extraction and concentration are required for point-of-care analysis in order to prevent an epidemic/pandemic disease outbreak. Typical silica-based NA extraction methods have limitations such as being time-consuming, requiring human intervention, and resulting in a low recovery yield. In this study, we have developed a pathogenic DNA extraction device based on electrokinetic separation incorporated with a silicon nitride (SiNx) nanofilter, which expedites the DNA extraction procedure with advantages of being convenient, efficient, and inexpensive. This DNA extraction device consists of a computer numerical control (CNC) milled-Teflon gadget with a cis-chamber as a cell lysate reservoir and a trans-chamber as a elution solution reservoir, with the SiNx nanofilter being inserted between the two chambers. The SiNx nanofilter was fabricated using a photolithographic method in conjunction with nanoimprinting. Approximately 7.2 million nanopores of 220 nm diameter were located at the center of the nanofilter. When a DC electric field is applied through the nanopores, DNA is transferred from the cis-chamber to the trans-chamber to isolate the DNA from the cell debris. To demonstrate the DNA extraction performance, we measured the absorbances at 260 and 280 nm and performed a real-time polymerase chain reaction (real-time PCR) using the recovered DNA to verify its feasibility for downstream genetic analysis. Moreover, the DNA extraction device was successfully operated using a 1.5 V alkaline battery, which verifies the portability of the device for point-of-care testing. Such an advanced DNA extraction system can be utilized in various fields including clinical analysis, pathogen detection, forensic analysis, and on-site detection.
Asunto(s)
ADN Bacteriano , Electroforesis/instrumentación , Escherichia coli , Sistemas de Atención de Punto , Reacción en Cadena en Tiempo Real de la Polimerasa , ADN Bacteriano/análisis , ADN Bacteriano/aislamiento & purificaciónRESUMEN
In this study, we used response surface methodology (RSM) to optimize the extraction conditions for recovering lutein from paprika leaves using accelerated solvent extraction (ASE). The lutein content was quantitatively analyzed using a UPLC equipped with a BEH C18 column. A central composite design (CCD) was employed for experimental design to obtain the optimized combination of extraction temperature (°C), static time (min), and solvent (EtOH, %). The experimental data obtained from a twenty sample set were fitted to a second-order polynomial equation using multiple regression analysis. The adjusted coefficient of determination (R(2)) for the lutein extraction model was 0.9518, and the probability value (p=0.0000) demonstrated a high significance for the regression model. The optimum extraction conditions for lutein were temperature: 93.26°C, static time: 5 min, and solvent: 79.63% EtOH. Under these conditions, the predicted extraction yield of lutein was 232.60 µg/g.
Asunto(s)
Antioxidantes/análisis , Capsicum/química , Luteína/análisis , Extractos Vegetales/química , Hojas de la Planta/química , Solventes/química , Antioxidantes/farmacología , Benzotiazoles/química , Cromatografía Liquida/métodos , Luteína/farmacología , Análisis de Regresión , Ácidos Sulfónicos/químicaRESUMEN
UNLABELLED: We aimed to identify the optimum conditions for the extraction of lutein from green tea using accelerated solvent extraction, and achieve improved analytical resolution and sensitivity between lutein and zeaxanthin using an ultra performance liquid chromatography (UPLC) system. The optimized method employed 80% ethanol as the extraction solvent, 160 °C as the temperature, 2 static cycles, and 5 min of static time. In the validation of the UPLC method, recovery was found to be in the range approximately 93.73 to 108.79%, with a correlation coefficient of 0.9974 and a relative standard deviation of <9.29% in inter- and intraday precision analyses. Finally, the lutein contents of green tea and green tea by-products were measured as 32.67 ± 0.70 and 18.18 ± 0.68 mg/100g dw, respectively. Furthermore, we verified that green tea by-products, which are discarded after producing green tea beverages, might be used as a great resource for massive lutein production. PRACTICAL APPLICATION: We have demonstrated that the common problem of inadequate resolution between lutein and zeaxanthin during carotenoid analyses can be overcome by optimizing the combined techniques of accelerated solvent extraction and ultra performance liquid chromatography (UPLC). UPLC was highly effective for saving time, solvent, and labor, as well as providing better resolution. The results in this study demonstrated that green tea by-products could be used as new sources for industrial lutein production owing to their massive production during the extraction of green tea beverages.