Application of interpretable machine learning algorithms to predict distant metastasis in ovarian clear cell carcinoma.

BACKGROUND: Ovarian clear cell carcinoma (OCCC) represents a subtype of ovarian epithelial carcinoma (OEC) known for its limited responsiveness to chemotherapy, and the onset of distant metastasis significantly impacts patient prognoses. This study aimed to identify potential risk factors contributing to the occurrence of distant metastasis in OCCC. METHODS: Utilizing the Surveillance, Epidemiology, and End Results (SEER) database, we identified patients diagnosed with OCCC between 2004 and 2015. The most influential factors were selected through the application of Gaussian Naive Bayes (GNB) and Adaboost machine learning algorithms, employing a Venn test for further refinement. Subsequently, six machine learning (ML) techniques, namely XGBoost, LightGBM, Random Forest (RF), Adaptive Boosting (Adaboost), Support Vector Machine (SVM), and Multilayer Perceptron (MLP), were employed to construct predictive models for distant metastasis. Shapley Additive Interpretation (SHAP) analysis facilitated a visual interpretation for individual patient. Model validity was assessed using accuracy, sensitivity, specificity, positive predictive value, negative predictive value, F1 score, and the area under the receiver operating characteristic curve (AUC). RESULTS: In the realm of predicting distant metastasis, the Random Forest (RF) model outperformed the other five machine learning algorithms. The RF model demonstrated accuracy, sensitivity, specificity, positive predictive value, negative predictive value, F1 score, and AUC (95% CI) values of 0.792 (0.762-0.823), 0.904 (0.835-0.973), 0.759 (0.731-0.787), 0.221 (0.186-0.256), 0.974 (0.967-0.982), 0.353 (0.306-0.399), and 0.834 (0.696-0.967), respectively, surpassing the performance of other models. Additionally, the calibration curve's Brier Score (95%) for the RF model reached the minimum value of 0.06256 (0.05753-0.06759). SHAP analysis provided independent explanations, reaffirming the critical clinical factors associated with the risk of metastasis in OCCC patients. CONCLUSIONS: This study successfully established a precise predictive model for OCCC patient metastasis using machine learning techniques, offering valuable support to clinicians in making informed clinical decisions.

Alternate pulses of ultrasound and electricity enhanced electrochemical process for p-nitrophenol degradation.

A novel alternated ultrasonic and electric pulse enhanced electrochemical process was developed and used for investigating its effectiveness on the degradation of p-nitrophenol (PNP) in an aqueous solution. The impacts of pulse mode, pH, cell voltage, supporting electrolyte concentration, ultrasonic power and the initial concentration of PNP on the performance of PNP degradation were evaluated. Possible pathway of PNP degradation in this system was proposed based on the intermediates identified by GC-MS. Experimental results showed that 94.1% of PNP could be removed at 2h in the dual-pulse ultrasound enhanced electrochemical (dual-pulse US-EC) process at mild operating conditions (i.e., pulse mode of electrochemical pulse time (TEC)=50 ms and ultrasonic pulse time (T US)=100 ms, initial pH of 3.0, cell voltage of 10 V, Na2SO4 concentration of 0.05 M, ultrasonic powder of 48.8 W and initial concentration of PNP of 100mg/L), compared with 89.0%, 58.9%, 2.4% in simultaneous ultrasound enhanced electrochemical (US-EC) process, pulsed electrochemical (EC) process and pulsed ultrasound (US), respectively. Moreover, energy used in the dual-pulse US-EC process was reduced by 50.4% as compared to the US-EC process. The degradation of PNP in the pulsed EC process, US-EC process and dual-pulse process followed pseudo-first-order kinetics. Therefore, the dual-pulse US-EC process was found to be a more effective technique for the degradation of PNP and would have a promising application in wastewater treatment.

Degradation of nitrobenzene in aqueous solution by dual-pulse ultrasound enhanced electrochemical process.

The present work reports a novel dual-pulse ultrasound enhanced electrochemical degradation (US-ECD) process that synchronizes alternatively ultrasound pulses and potential pulses to degrade nitrobenzene in aqueous solution with a high percentage degradation and low energy consumption. In comparison to the test results generated from the conventional US-ECD and original electrochemical degradation (ECD) process, the dual-pulse US-ECD process increased degradation percentages to nitrobenzene by 2% and 17%, respectively, while energy used in the pulse process was only about 46.5% of that was used in the conventional US-ECD process. Test results demonstrated a superior performance of the dual-pulse US-ECD process over those of other conventional ones. Impacts of pulse mode, initial pH value, cell voltage, supporting electrolyte concentration and ultrasonic power on the process performances were investigated. With operation conditions optimized in the study at pH=3.0, cell voltage=10 V, ultrasonic power=48.84 W, electrolyte concentration=0.1M and an experiment running time of 30 min, the percentage degradation of nitrobenzene could reach 80% (US pulse time=50 ms and ECD pulse time=50 ms). This process provided a reliable and effective technical approach to degrade nitrobenzene in aqueous solution and significantly reduced energy consumption in comparison to the conventional US-ECD or original ECD treatment.

Ultrasonic recovery of copper and iron through the simultaneous utilization of Printed Circuit Boards (PCB) spent acid etching solution and PCB waste sludge.

A method was developed to recover the copper and iron from Printed Circuit Boards (PCB) manufacturing generated spent acid etching solution and waste sludge with ultrasonic energy at laboratory scale. It demonstrated that copper-containing PCB spent etching solution could be utilized as a leaching solution to leach copper from copper contained PCB waste sludge. It also indicated that lime could be used as an alkaline precipitating agent in this method to precipitate iron from the mixture of acidic PCB spent etching solution and waste sludge. This method provided an effective technique for the recovery of copper and iron through simultaneous use of PCB spent acid solution and waste sludge. The leaching rates of copper and iron enhanced with ultrasound energy were reached at 93.76% and 2.07% respectively and effectively separated copper from iron. Followed by applying lime to precipitate copper from the mixture of leachate and rinsing water produced by the copper and iron separation, about 99.99% and 1.29% of soluble copper and calcium were settled as the solids respectively. Furthermore the settled copper could be made as commercial rate copper. The process performance parameters studied were pH, ultrasonic power, and temperature. This method provided a simple and reliable technique to recover copper and iron from waste streams generated by PCB manufacturing, and would significantly reduce the cost of chemicals used in the recovery.

Shape-controlled synthesis of cuprous oxide nanocrystals via the electrochemical route with H2O-polyol mix-solvent and their behaviors of adsorption.

With H2O-polyols (glycerol, ethylene glycol and polyethylene glycol 400) mix-solvent as media in the NaCl-NaOH-NaNO3 electrolytic system, we developed a simple and efficient electrochemical route for the synthesis of Cu2O nanocrystals with different morphology. The SEM results indicated that electrolytic media and current density had a great influence on the shape of Cu2O crystals. The mono-disperse and uniform sphere Cu2O nanoparticles could be readily obtained in H2O-glycerol (or H2O-ethylene glycol) mix-solvent (1:1 volume ratio) and current density 5 mA/cm2. The sphere Cu2O nanoparticles exhibit excellent adsorption ability for organic dyes (methyl orange, fuchsin acid and methyl blue), which is obviously superior to that of the irregular shape Cu2O crystals prepared in H2O-PEG400 (polyethylene glycol 400) mix-solvent. The present work further confirmed that the adsorption ability of Cu2O crystals was related to their size and shape.

Ultra-long periodically-twinned copper nanowires evolving from dendrites in interfacial electrodeposition.

A simple and efficient template-free electrochemical route has been utilized to prepare ultra-long Cu nanowires at a steady water-oleic acid interface. The as-prepared nanowires have uniform diameters and aspect ratios of more than 1000, and periodically twinned structure. The investigation of the growth process has indicated that these nanowires actually evolve from the branches of dendrites. The morphology of the products can be tuned by controlling experimental parameters, such as electrodeposition time, metallic ions concentration, current density, the geometry of electric field and the property and state of the interfaces. The experimental results show that the steady water-oleic acid interface and a parallel electric field are crucial for the formation of dendritic Cu nanowires. It is believed that the higher Cu(2+) concentration near the water-oleic acid interface is favorable to induce interfacial electrodeposition. The electrochemical route is also suitable for the synthesis of other metal nanowires, such as Co nanowires.

Multiple heavy metals extraction and recovery from hazardous electroplating sludge waste via ultrasonically enhanced two-stage acid leaching.

An ultrasonically enhanced two-stage acid leaching process on extracting and recovering multiple heavy metals from actual electroplating sludge was studied in lab tests. It provided an effective technique for separation of valuable metals (Cu, Ni and Zn) from less valuable metals (Fe and Cr) in electroplating sludge. The efficiency of the process had been measured with the leaching efficiencies and recovery rates of the metals. Enhanced by ultrasonic power, the first-stage acid leaching demonstrated leaching rates of 96.72%, 97.77%, 98.00%, 53.03%, and 0.44% for Cu, Ni, Zn, Cr, and Fe respectively, effectively separated half of Cr and almost all of Fe from mixed metals. The subsequent second-stage leaching achieved leaching rates of 75.03%, 81.05%, 81.39%, 1.02%, and 0% for Cu, Ni, Zn, Cr, and Fe that further separated Cu, Ni, and Zn from mixed metals. With the stabilized two-stage ultrasonically enhanced leaching, the resulting over all recovery rates of Cu, Ni, Zn, Cr and Fe from electroplating sludge could be achieved at 97.42%, 98.46%, 98.63%, 98.32% and 100% respectively, with Cr and Fe in solids and the rest of the metals in an aqueous solution discharged from the leaching system. The process performance parameters studied were pH, ultrasonic power, and contact time. The results were also confirmed in an industrial pilot-scale test, and same high metal recoveries were performed.

The ultrasonically assisted metals recovery treatment of printed circuit board waste sludge by leaching separation.

This paper provides a practical technique that realized industrial scale copper and iron separation from printed circuit board (PCB) waste sludge by ultrasonically assisted acid leaching in a low cost, low energy consumption and zero discharge of wastes manner. The separation efficiencies of copper and iron from acid leaching with assistance of ultrasound were compared with the one without assistance of ultrasound and the effects of the leaching procedure, pH value, and ultrasonic strength have been investigated in the paper. With the appropriate leaching procedure, a final pH of 3.0, an ultrasonic generator power of 160 W (in 1l tank), leaching time of 60 min, leaching efficiencies of copper and iron had reached 97.83% and 1.23%, respectively. Therefore the separation of copper and iron in PCB waste sludge was virtually achieved. The lab results had been successfully applied to the industrial scaled applications in a heavy metal recovery plant in city of Huizhou, China for more than two years. It has great potentials to be used in even the broad metal recovery practices.

A new fungicide produced by a Streptomyces sp. GAAS7310.

Directed bioassay guided fraction led to a new macrodiolide antimycin A(17) (1), isolated from a Streptomyces sp. GAAS7310, which showed significant antimicrobial activity against eleven fungal species, including Curvularia lunata (Wakker) Boed, Rhizopus nigrtcans Ehrb and Colletotrichum nigrum EL. et Halst. The structure was unambiguously established by interpretation of 1D and 2D NMR data and comparison with the known antimycin A(1a).