Development and validation of high definition phenotype-based mortality prediction in critical care units.

OBJECTIVES: The objectives of this study are to construct the high definition phenotype (HDP), a novel time-series data structure composed of both primary and derived parameters, using heterogeneous clinical sources and to determine whether different predictive models can utilize the HDP in the neonatal intensive care unit (NICU) to improve neonatal mortality prediction in clinical settings. MATERIALS AND METHODS: A total of 49 primary data parameters were collected from July 2018 to May 2020 from eight level-III NICUs. From a total of 1546 patients, 757 patients were found to contain sufficient fixed, intermittent, and continuous data to create HDPs. Two different predictive models utilizing the HDP, one a logistic regression model (LRM) and the other a deep learning long-short-term memory (LSTM) model, were constructed to predict neonatal mortality at multiple time points during the patient hospitalization. The results were compared with previous illness severity scores, including SNAPPE, SNAPPE-II, CRIB, and CRIB-II. RESULTS: A HDP matrix, including 12 221 536 minutes of patient stay in NICU, was constructed. The LRM model and the LSTM model performed better than existing neonatal illness severity scores in predicting mortality using the area under the receiver operating characteristic curve (AUC) metric. An ablation study showed that utilizing continuous parameters alone results in an AUC score of >80% for both LRM and LSTM, but combining fixed, intermittent, and continuous parameters in the HDP results in scores >85%. The probability of mortality predictive score has recall and precision of 0.88 and 0.77 for the LRM and 0.97 and 0.85 for the LSTM. CONCLUSIONS AND RELEVANCE: The HDP data structure supports multiple analytic techniques, including the statistical LRM approach and the machine learning LSTM approach used in this study. LRM and LSTM predictive models of neonatal mortality utilizing the HDP performed better than existing neonatal illness severity scores. Further research is necessary to create HDP-based clinical decision tools to detect the early onset of neonatal morbidities.

Designing a bed-side system for predicting length of stay in a neonatal intensive care unit.

Increased length of stay (LOS) in intensive care units is directly associated with the financial burden, anxiety, and increased mortality risks. In the current study, we have incorporated the association of day-to-day nutrition and medication data of the patient during its stay in hospital with its predicted LOS. To demonstrate the same, we developed a model to predict the LOS using risk factors (a) perinatal and antenatal details, (b) deviation of nutrition and medication dosage from guidelines, and (c) clinical diagnoses encountered during NICU stay. Data of 836 patient records (12 months) from two NICU sites were used and validated on 211 patient records (4 months). A bedside user interface integrated with EMR has been designed to display the model performance results on the validation dataset. The study shows that each gestation age group of patients has unique and independent risk factors associated with the LOS. The gestation is a significant risk factor for neonates < 34 weeks, nutrition deviation for < 32 weeks, and clinical diagnosis (sepsis) for ≥ 32 weeks. Patients on medications had considerable extra LOS for ≥ 32 weeks' gestation. The presented LOS model is tailored for each patient, and deviations from the recommended nutrition and medication guidelines were significantly associated with the predicted LOS.

Machine Learning-Based Automatic Classification of Video Recorded Neonatal Manipulations and Associated Physiological Parameters: A Feasibility Study.

Our objective in this study was to determine if machine learning (ML) can automatically recognize neonatal manipulations, along with associated changes in physiological parameters. A retrospective observational study was carried out in two Neonatal Intensive Care Units (NICUs) between December 2019 to April 2020. Both the video and physiological data (heart rate (HR) and oxygen saturation (SpO2)) were captured during NICU hospitalization. The proposed classification of neonatal manipulations was achieved by a deep learning system consisting of an Inception-v3 convolutional neural network (CNN), followed by transfer learning layers of Long Short-Term Memory (LSTM). Physiological signals prior to manipulations (baseline) were compared to during and after manipulations. The validation of the system was done using the leave-one-out strategy with input of 8 s of video exhibiting manipulation activity. Ten neonates were video recorded during an average length of stay of 24.5 days. Each neonate had an average of 528 manipulations during their NICU hospitalization, with the average duration of performing these manipulations varying from 28.9 s for patting, 45.5 s for a diaper change, and 108.9 s for tube feeding. The accuracy of the system was 95% for training and 85% for the validation dataset. In neonates <32 weeks' gestation, diaper changes were associated with significant changes in HR and SpO2, and, for neonates ≥32 weeks' gestation, patting and tube feeding were associated with significant changes in HR. The presented system can classify and document the manipulations with high accuracy. Moreover, the study suggests that manipulations impact physiological parameters.

Development of data dictionary for neonatal intensive care unit: advancement towards a better critical care unit.

BACKGROUND: Critical care units (CCUs) with extensive use of various monitoring devices generate massive data. To utilize the valuable information of these devices; data are collected and stored using systems like clinical information system and laboratory information management system. These systems are proprietary, allow limited access to their database and, have the vendor-specific clinical implementation. In this study, we focus on developing an open-source web-based meta-data repository for CCU representing stay of the patient with relevant details. METHODS: After developing the web-based open-source repository named data dictionary (DD), we analyzed prospective data from 2 sites for 4 months for data quality dimensions (completeness, timeliness, validity, accuracy, and consistency), morbidity, and clinical outcomes. We used a regression model to highlight the significance of practice variations linked with various quality indicators. RESULTS: DD with 1555 fields (89.6% categorical and 11.4% text fields) is presented to cover the clinical workflow of a CCU. The overall quality of 1795 patient days data with respect to standard quality dimensions is 87%. The data exhibit 88% completeness, 97% accuracy, 91% timeliness, and 94% validity in terms of representing CCU processes. The data scores only 67% in terms of consistency. Furthermore, quality indicators and practice variations are strongly correlated (P < 0.05). CONCLUSION: This study documents DD for standardized data collection in CCU. DD provides robust data and insights for audit purposes and pathways for CCU to target practice improvements leading to specific quality improvements.

Exploring electrochemistry of carbon nanodots and its application in noninvasive bacterial growth monitoring.

Cost effective and miniaturized methods aiming for high throughput monitoring of bacterial growth are of great significance, especially for tracking disease progression in early stage as well as in screening antibiotic resistant species. Here, we demonstrate an electrochemical platform for noninvasive monitoring of bacterial growth by encapsulating bacterial cells and carbon nanodots in alginate microspheres. The synthesized carbon nanodots have been explored for electrochemical properties, and its redox properties have been utilized for developing bacterial growth monitoring platform. These synthesized CDs are sensitive to pH change and respond as change in redox potential over time as pH of the medium changes due to growth and metabolic activities of bacteria. We determined the bacterial growth kinetics by measuring the redox potential changes of the carbon nanodots over time. The developed platform has been demonstrated to detect the presence of bacteria, the difference in growth rates of bacteria and its susceptibility to the antibiotic with low bacterial counts (103 CFU) in 20 min; thus, redox properties of CDs has the potential to provide a sensitive detection platform.

Structural switching electrochemical DNA aptasensor for the rapid diagnosis of tuberculous meningitis.

BACKGROUND: Tuberculous meningitis (TBM) is the most devastating manifestation of extra-pulmonary tuberculosis. About 33% of TBM patients die due to very late diagnosis of the disease. Conventional diagnostic methods based on signs and symptoms, cerebrospinal fluid (CSF) smear microscopy or liquid culture suffer from either poor sensitivity or long turnaround time (up to 8 weeks). Therefore, in order to manage the disease efficiently, there is an urgent and unmet need for a rapid and reliable diagnostic test. METHODS: In the current study, to address the diagnostic challenge of TBM, a highly rapid and sensitive structural switching electrochemical aptasensor was developed by combining the electrochemical property of methylene blue (MB) with the molecular recognition ability of a ssDNA aptamer. To demonstrate the clinical diagnostic utility of the developed aptasensor, a blinded study was performed on 81 archived CSF specimens using differential pulse voltammetry. RESULTS: The electrochemical aptasensor developed in the current study can detect as low as 10 pg HspX in CSF background and yields a highly discriminatory response (P<0.0001) for TBM and not-TBM categories with ~95% sensitivity and ~97.5% specificity and has the ability to deliver sample-to-answer in ≤30 minutes. CONCLUSION: In summary, we demonstrate a new aptamer-based electrochemical biosensing strategy by exploiting the target-induced structural switching of H63 SL-2 M6 aptamer and electroactivity of aptamer-tagged MB for the detection of HspX in CSF samples for the diagnosis of TBM. Further, the clinical utility of this sensor could be extended for the diagnosis of other forms of tuberculosis in the near future.

Aptamer-mediated colorimetric and electrochemical detection of Pseudomonas aeruginosa utilizing peroxidase-mimic activity of gold NanoZyme.

Despite of various advancements in biosensing, a rapid, accurate, and on-site detection of a bacterial pathogen is a real challenge due to the lack of appropriate diagnostic platforms. To address this unmet need, we herein report an aptamer-mediated tunable NanoZyme sensor for the detection of Pseudomonas aeruginosa, an infectious bacterial pathogen. Our approach exploits the inherent peroxidase-like NanoZyme activity of gold nanoparticles (GNPs) in combination with high affinity and specificity of a Pseudomonas aeruginosa-specific aptamer (F23). The presence of aptamer inhibits the inherent peroxidase-like activity of GNPs by simple adsorption on to the surface of GNPs. However, in the presence of cognate target (P. aeruginosa), owing to the high affinity for P. aeruginosa, the aptamer leaves the GNP surface, allowing GNPs to resume their peroxidase-like activity, resulting in oxidation of 3,3',5,5'-tetramethylbenzidine (TMB). As TMB is an electrochemically active species, we have been able to translate the NanoZyme-based method into an ultrasensitive electrochemical assay using disposable carbon screen-printed electrode. This approach is highly sensitive and allows us to rapidly detect P. aeruginosa with a low-end detection limit of ~ 60 CFU/mL in water within 10 min. This generic aptamer-NanoZyme-based electrochemical sensing strategy may, in principle, be applicable for the detection of various other bacterial pathogens.

Aptamer-Based TB Antigen Tests for the Rapid Diagnosis of Pulmonary Tuberculosis: Potential Utility in Screening for Tuberculosis.

Pulmonary tuberculosis is the most common manifestation of tuberculosis, and to this day, sputum smear microscopy remains the most widely used diagnostic test in resource-limited settings despite its suboptimal sensitivity. Here we report the development of two DNA aptamer-based diagnostic tests, namely aptamer linked immobilized sorbent assay (Aptamer ALISA) and electrochemical sensor (ECS), for the direct detection of a TB biomarker HspX in sputum. First we compared the performance of Aptamer ALISA with anti-HspX polyclonal antibody-based enzyme linked immunosorbent assay (Antibody ELISA) in a blinded study of 314 sputum specimens. Aptamer ALISA displayed a high sensitivity of 94.1% (95% CI 86.8-98%) as compared to 68.2% sensitivity (95% CI 57.2-77.9%) of Antibody ELISA ( p-value < 0.05) using culture as the reference standard without compromising test specificity of 100%. Out of nine smear-negative culture-positive samples, six were positive by Aptamer ALISA and only two were detected by Antibody ELISA. ALISA detected as positive 80 of 85 culture-positive TB as compared to 57 of 81 diagnosed as TB by X-ray ( p-value < 0.0001). These findings demonstrate the superiority of the aptamer-based test over smear microscopy, antibody-based ELISA, and chest X-ray for TB detection ( p-value < 0.0001 for all). Further, we have developed a ∼30 min point-of-care ECS test that discriminates between tuberculous and nontuberculous sputum with a sensitivity of ∼92.3% and specificity of 91.2%. The tests developed in the current study cost ∼$1-3/test and have potential utility in active case finding in high-risk groups and screening for pulmonary TB among presumptive TB subjects.

Rapid Detection Device for Salmonella typhi in Milk, Juice, Water and Calf Serum.

A limit of detection of 200 CFU/mL of Salmonella typhi spiked in various sample matrices were achieved in 30 min. The sample matrices were raw/unprocessed milk, commercially available milk, juice from packed bottles, fresh juice from carts, potable water, turbid water and calf serum. The complete protocol comprised of three steps: (a) cell lysis (b) nucleic acid amplification and (c) an in situ optical detection. The cell lysis was carried out using a simple heating based protocol, while the loop-mediated isothermal amplification of DNA was carried out by an in-house designed and fabricated system. The developed system consists of an aluminum block fitted with two cartridge heaters along with a thermocouple. The system was coupled to a light source and spectrometer for a simultaneous in situ detection. Primers specific for STY2879 gene were used to amplify the nucleic acid sequence, isolated from S. typhi cells. The protocol involves 15 min of cell lysis and DNA isolation followed by 15 min for isothermal amplification and simultaneous detection. No cross-reactivity of the primers were observed at 106 CFU/mL of Escherichia coli, Vibrio cholerae, Salmonella typhimurium, Salmonella paratyphi A, Pseudomonas aeruginosa, Bacillus cereus, Lysteria monocytogenes, Clostridium botulinum, Staphylococcus aureus and Salmonella havana. In addition, the system was able to detect S. typhi of 200 CFU/mL in a concoction of 106 CFU/mL of E. coli, 106 CFU/mL of V. cholerae, and 106 CFU/mL of hepatocyte-derived cellular carcinoma HUH7 cells. The proposed rapid diagnostic system shows a promising future in the field of food and medical diagnostics.

Electrochemical DNA sensor for anthrax toxin activator gene atxA-detection of PCR amplicons.

We report the DNA probe functionalized electrochemical genosensor for the detection of Bacillus anthracis, specific towards the regulatory gene atxA. The DNA sensor is fabricated on electrochemically deposited gold nanoparticle on self assembled layer of (3-Mercaptopropyl) trimethoxysilane (MPTS) on GC electrode. DNA hybridization is monitored by differential pulse voltammogram (DPV). The modified GC electrode is characterized by atomic force microscopy (AFM), cyclic voltammetry (CV), and electrochemical impedance spectroscopy (EIS) method. We also quantified the DNA probe density on electrode surface by the chronocoulometric method. The detection is specific and selective for atxA gene by DNA probe on the electrode surface. No report is available for the detection of B. anthracis by using atxA an anthrax toxin activator gene. In the light of real and complex sample, we have studied the PCR amplicons of 303, 361 and 568 base pairs by using symmetric and asymmetric PCR approaches. The DNA probe of atxA gene efficiently hybridizes with different base pairs of PCR amplicons. The detection limit is found to be 1.0 pM (S/N ratio=3). The results indicate that the DNA sensor is able to detect synthetic target as well as PCR amplicons of different base pairs.

An electrochemical genosensor for Salmonella typhi on gold nanoparticles-mercaptosilane modified screen printed electrode.

In this work, we fabricated a system of integrated self-assembled layer of organosilane 3-mercaptopropyltrimethoxy silane (MPTS) on the screen printed electrode (SPE) and electrochemically deposited gold nanoparticle for Salmonella typhi detection employing Vi gene as a molecular marker. Thiolated DNA probe was immobilized on a gold nanoparticle (AuNP) modified SPE for DNA hybridization assay using methylene blue as redox (electroactive) hybridization indicator, and signal was monitored by differential pulse voltammetry (DPV) method. The modified SPE was characterized by cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and atomic force microscopy (AFM) method. The DNA biosensor showed excellent performances with high sensitivity and good selectivity. The current response was linear with the target sequence concentrations ranging from 1.0 × 10(-11) to 0.5 × 10(-8)M and the detection limit was found to be 50 (± 2.1)pM. The DNA biosensor showed good discrimination ability to the one-base, two-base and three-base mismatched sequences. The fabricated genosensor could also be regenerated easily and reused for three to four times for further hybridization studies.