RESUMO
We developed an interpretable machine learning algorithm that prospectively predicts the risk of thrombocytopenia in older critically ill patients during their stay in the intensive care unit (ICU), ultimately aiding clinical decision-making and improving patient care. Data from 2286 geriatric patients who underwent surgery and were admitted to the ICU of Dongyang People's Hospital between 2012 and 2021 were retrospectively analyzed. Integrated algorithms were developed, and four machine-learning algorithms were used. Selected characteristics included common demographic data, biochemical indicators, and vital signs. Eight key variables were selected using the Least Absolute Shrinkage and Selection Operator and Random Forest Algorithm. Thrombocytopenia occurred in 18.2% of postoperative geriatric patients, with a higher mortality rate. The C5.0 model showed the best performance, with an area under the receiver operating characteristic curve close to 0.85, along with unparalleled accuracy, precision, specificity, recall, and balanced accuracy scores of 0.88, 0.98, 0.89, 0.98, and 0.85, respectively. The support vector machine model excelled at predictively assessing thrombocytopenia severity, demonstrating an accuracy rate of 0.80 in the MIMIC database. Thus, our machine learning-based models have considerable potential in effectively predicting the risk and severity of postoperative thrombocytopenia in geriatric ICU patients for better clinical decision-making and patient care.
Assuntos
Unidades de Terapia Intensiva , Aprendizado de Máquina , Trombocitopenia , Humanos , Idoso , Feminino , Masculino , Estudos Retrospectivos , Idoso de 80 Anos ou mais , Algoritmos , Estado Terminal , Curva ROC , Tomada de Decisão ClínicaRESUMO
Acute kidney injury (AKI) is a common postoperative complication among patients in the neurological intensive care unit (NICU), often resulting in poor prognosis and high mortality. In this retrospective cohort study, we established a model for predicting AKI following brain surgery based on an ensemble machine learning algorithm using data from 582 postoperative patients admitted to the NICU at the Dongyang People's Hospital from March 1, 2017, to January 31, 2020. Demographic, clinical, and intraoperative data were collected. Four machine learning algorithms (C5.0, support vector machine, Bayes, and XGBoost) were used to develop the ensemble algorithm. The AKI incidence in critically ill patients after brain surgery was 20.8%. Intraoperative blood pressure; postoperative oxygenation index; oxygen saturation; and creatinine, albumin, urea, and calcium levels were associated with the postoperative AKI occurrence. The area under the curve value for the ensembled model was 0.85. The accuracy, precision, specificity, recall, and balanced accuracy values were 0.81, 0.86, 0.44, 0.91, and 0.68, respectively, indicating good predictive ability. Ultimately, the models using perioperative variables exhibited good discriminatory ability for early prediction of postoperative AKI risk in patients admitted to the NICU. Thus, the ensemble machine learning algorithm may be a valuable tool for forecasting AKI.
Assuntos
Injúria Renal Aguda , Humanos , Estudos Retrospectivos , Teorema de Bayes , Injúria Renal Aguda/diagnóstico , Injúria Renal Aguda/epidemiologia , Injúria Renal Aguda/etiologia , Aprendizado de Máquina , Algoritmos , EncéfaloRESUMO
A novel transparent conductive conductor composed of a silver nanowire (AgNW) network and MoOx on a flexible polyethylene terephthalate (PET) substrate, with contemporaneously improved adhesion and reduced resistivity, is prepared using the full-solution process without high-temperature annealing. Under the optimized conditions, a MoOx/AgNW/MoOx multilayer is achieved, which shows much superior optoelectronic performance to that obtained from ITO with a high optical transmittance of 89.2% and a low sheet resistance of â¼12.5 Ω/sq. Unlike pure AgNW films, the sheet resistance is little changed after the tape and ultrasonication tests, illustrating a very strong adhesion to the PET substrate after the encapsulation of MoOx. Moreover, the multilayer film exhibits excellent stability to resist mechanical bending and acid damage. In addition, the successful implementation of the flexible transparent heater demonstrates the practical application value of the electrode.
RESUMO
Copper nanowires (CuNWs) are remarkable components that can replace indium tin oxide as transparent electrodes due to their low cost, high conductivity and acceptable transmittance. However, a common coating method can cause poor electrical, optical and adhesive properties because of the creation of loosely connected junctions. In addition, the unsatisfactory thermal and environmental stabilities limit the practical applications. These problems should be overcome in CuNW-based films for reliable transparent electrodes through material and engineering approaches. In this work, a novel transparent composite electrode composed of chitosan and CuNWs on a flexible polyethylene terephthalate (PET) substrate, with synchronously strengthened adhesion, as well as heightened transmittance, reduced resistivity, improved flexibility, enhanced thermal stability and increased environmental stability, was prepared without vacuum processing and high-temperature annealing. The effects of the number of CuNW network layers and chitosan concentration on the performance of chitosan/CuNW composite transparent electrodes were studied. The resulting electrodes exhibitan excellent conductivity (sheet resistance: 15.6 Ω sq-1) and a superior optical transmittance (â¼87%) at 550 nm. Calculation of the figure of merit displays a high value of 168, which is the highest among all the reported CuNW-based transparent electrodes. Meanwhile, the sheet resistance did not show great change after 10 tape tests and 10 000 bending cycles, suggesting good adhesion to the PET substrate and outstanding mechanical flexibility. Moreover, the composite transparent electrodes show good stability to resist long-term storage and temperature variation in thermal environment.
RESUMO
Outer space is a challenging environment for all forms of life, and dormant spores of bacteria have been frequently used to study the survival of terrestrial life in a space journey. Previous work showed that outer space vacuum alone can kill bacterial spores. However, the responses and mechanisms of resistance of individual spores to space vacuum are unclear. Here, we examined spores' molecular changes under simulated space vacuum (~10-5 Pa) using micro-Raman spectroscopy and found that this vacuum did not cause significant denaturation of spore protein. Then, live-cell microscopy was developed to investigate the temporal events during germination, outgrowth, and growth of individual Bacillus spores. The results showed that after exposure to simulated space vacuum for 10 days, viability of spores of two Bacillus species was reduced up to 35%, but all spores retained their large Ca2+-dipicolinic acid depot. Some of the killed spores did not germinate, and the remaining germinated but did not proceed to vegetative growth. The vacuum treatment slowed spore germination, and changed average times of all major germination events. In addition, viable vacuum-treated spores exhibited much greater sensitivity than untreated spores to dry heat and hyperosmotic stress. Among spores' resistance mechanisms to high vacuum, DNA-protective α/ß-type small acid-soluble proteins, and non-homologous end joining and base excision repair of DNA played the most important roles, especially against multiple cycles of vacuum treatment. Overall, these results give new insight into individual spore's responses to space vacuum and provide new techniques for microorganism analysis at the single-cell level.
RESUMO
DNA damage kills dry-heated spores of Bacillus subtilis, but dry-heat-treatment effects on spore germination and outgrowth have not been studied. This is important, since if dry-heat-killed spores germinate and undergo outgrowth, toxic proteins could be synthesized. Here, Raman spectroscopy and differential interference contrast microscopy were used to study germination and outgrowth of individual dry-heat-treated B. subtilis and Bacillus megaterium spores. The major findings in this work were as follows: (i) spores dry-heat-treated at 140°C for 20 min lost nearly all viability but retained their Ca2+-dipicolinic acid (CaDPA) depot; (ii) in most cases, dry-heat treatment increased the average times and variability of all major germination events in B. subtilis spore germination with nutrient germinants or CaDPA, and in one nutrient germination event with B. megaterium spores; (iii) B. subtilis spore germination with dodecylamine, which activates the spore CaDPA release channel, was unaffected by dry-heat treatment; (iv) these results indicate that dry-heat treatment likely damages spore proteins important in nutrient germinant recognition and cortex peptidoglycan hydrolysis, but not CaDPA release itself; and (v) analysis of single spores incubated on nutrient-rich agar showed that while dry-heat-treated spores that are dead can complete germination, they cannot proceed into outgrowth and thus not to vegetative growth. The results of this study provide new information on the effects of dry heat on bacterial spores and indicate that dry-heat sterilization regimens should produce spores that cannot outgrow and thus cannot synthesize potentially dangerous proteins.IMPORTANCE Much research has shown that high-temperature dry heat is a promising means for the inactivation of spores on medical devices and spacecraft decontamination. Dry heat is known to kill Bacillus subtilis spores by DNA damage. However, knowledge about the effects of dry-heat treatment on spore germination and outgrowth is limited, especially at the single spore level. In the current work, Raman spectroscopy and differential interference contrast microscopy were used to analyze CaDPA levels in and kinetics of nutrient- and non-nutrient germination of multiple individual dry-heat-treated B. subtilis and Bacillus megaterium spores that were largely dead. The outgrowth and subsequent cell division of these germinated but dead dry-heat-treated spores were also examined. The knowledge obtained in this study will help understand the effects of dry heat on spores both on Earth and in space, and indicates that dry heat can be safely used for sterilization purposes.
Assuntos
Bacillus megaterium/fisiologia , Bacillus subtilis/fisiologia , Temperatura Alta , Esporos Bacterianos/fisiologia , Esterilização , Cinética , Microscopia de Interferência , Análise Espectral RamanRESUMO
Transparent conductive multilayer thin films of silver (Ag)-embedded barium stannate (BaSnO3) structures have been deposited onto flexible polycarbonate substrates by magnetron sputtering at room temperature to develop an indium free transparent flexible electrode. The effect of thicknesses of Ag mid-layer and barium stannate layers on optical and electrical properties were investigated, and the mechanisms of conduction and transmittance were discussed. The highest value of figure of merit is 25.5 × 10-3 Ω-1 for the BaSnO3/Ag/BaSnO3 multilayer flexible thin films with 9 nm thick silver mid-layer and 50 nm thick barium stannate layers, while the average optical transmittance in the visible range from 380 to 780 nm is above 87%, the resistivity is 9.66 × 10-5 Ω · cm, and the sheet resistance is 9.89 Ω/sq. The change rate of resistivity is under 10% after repeated bending of the multilayer flexible thin films. These results indicate that Ag-based barium stannate multilayer flexible thin films can be used as transparent flexible electrodes in various flexible optoelectronic devices.
RESUMO
Optical manipulation and label-free characterization of nanoscale structures open up new possibilities for assembly and control of nanodevices and biomolecules. Optical tweezers integrated with Raman spectroscopy allows analyzing a single trapped particle, but is generally less effective for individual nanoparticles. The main challenge is the weak gradient force on nanoparticles that is insufficient to overcome the destabilizing effect of scattering force and Brownian motion. Here, we present standing-wave Raman tweezers for stable trapping and sensitive characterization of single isolated nanostructures with a low laser power by combining a standing-wave optical trap with confocal Raman spectroscopy. This scheme has stronger intensity gradients and balanced scattering forces, and thus can be used to analyze many nanoparticles that cannot be measured with single-beam Raman tweezers, including individual single-walled carbon nanotubes (SWCNT), graphene flakes, biological particles, SERS-active metal nanoparticles, and high-refractive semiconductor nanoparticles. This would enable sorting and characterization of specific SWCNTs and other nanoparticles based on their increased Raman fingerprints.
RESUMO
The BaSnO3/Ag/BaSnO3 multilayer structure was designed and fabricated on a quartz glass by magnetron sputtering, followed by an annealing process at a temperature from 150 to 750 °C in air. In this paper, we investigated the influence of the annealing temperature on the structural, optical, and electrical properties of the multilayers and proposed the mechanisms of conduction and transmittance. The maximum value of the figure of merit of 31.8 × 10(-3) Ω(-1) was achieved for the BaSnO3/Ag/BaSnO3 multilayer thin films annealed at 150 °C, while the average optical transmittance in the visible ranges was >84 %, the resistivity was 5.71 × 10(-5) Ω cm, and the sheet resistance was 5.57 Ω/sq. When annealed at below 600 °C, the values of resistivity and transmittance of the multilayers were within an acceptable range (resistivity <5.0 × 10(-4) Ω cm, transmittance >80 %). The observed property of the multilayer film is suitable for the application of transparent conductive electrodes.