Machine learning-based prediction of long-term mortality in patients with atrial fibrillation and coronary heart disease aged 60 years and over / 中华老年医学杂志

Objective:To establish a long-term mortality rate prediction model for patients aged 60 years and over with atrial fibrillation and coronary heart disease using the machine learning method, and identify the corresponding risk factors of mortality.Methods:In this retrospective cohort study, a total of 329(11 cases lost of follow-up)patients with 183 males(55.6%)and 146 females(44.4%), aged(77.8±7.3)years, and 142 patients aged 80 years or older(43.2%)were selected in our hospitals from January 2013 to March 2015.And their clinical data on atrial fibrillation and coronary heart disease were analyzed.They were divided into the death group(151 cases)and the survival group(167 cases)according to the survival outcome.In addition, 60 patients aged 60 years and over admitted to our hospitals from April to July 2015 with atrial fibrillation and coronary heart disease were selected as external data validation set.The clinical data included age, gender, body mass index, diagnosis, co-morbidity, laboratory indicators, electrocardiogram, echocardiogram, treatment data.These patients were followed up for at least 6 years, and the main adverse cardiovascular and cerebrovascular events(MACCE), including death, were recorded.Finally, the data of the enrolled patients were randomly divided into the training set and the test set according to the ratio of 9∶1, Different models were established to predict the long-term mortality of patients with atrial fibrillation and coronary heart disease by machine learning algorithm.The optimal model was established by substituting external data(60 cases)into the model for verification and comparison.The top 20 risk factors for mortality were determined by Shapley additive explanation(SHAP)algorithm.Results:A total of 329 hospitalized patients were included in this study, the overall median follow-up time was 77.0 months(95% CI: 54.0~84.0), 11 cases lost during follow-up(3.3%), and 151 cases died(45.9%). The analysis found that the areas under the ROC curve for a support vector machine(SVM)model, k-Nearest Neighbor(KNN)model, decision tree model, random forest model, ADABoost model, XGBoost model and logistic regression model were 0.76, 0.75, 0.75, 0.91, 0.86, 0.85 and 0.81, respectively.The random forest model had the highest prediction efficiency, with the accuracy of 0.789 and F1 value of 0.806, which was better than the logistic regression model[the Area Under Receiver Operating Characteristic Curve(AUC): 0.91 vs.0.81, P<0.05]. D-dimer, age, number of MACCE, left ventricular ejection fraction, serum albumin level, anemia, New York Heart Association(NYHA)grade, history of old myocardial infarction, estimated glomerular filtration rate(eGFR)and resting heart rate were important risk factors for predicting long-term mortality. Conclusions:The random forest model based on machine learning method can predict the long-term mortality of patients with atrial fibrillation and coronary heart disease aged 60 years and over, have a good identification ability.Its accuracy is higher than that of the traditional Logistic regression model.Reducing the long-term mortality and improving the long-term outcomes can be achieved by intervening on D-dimer levels, correcting hypoproteinemia and anemia, improving cardiac function and controlling resting ventricular rates.

In vivo characterization of a novel Cerasome based multi-modality imaging probe / 中华核医学与分子影像杂志

Objective To prepare a novel dual-modality imaging probe based on Cerasome nano-materials, and evaluate its in vivo biodistribution and pharmacokinetic properties. Methods ICG encapsu-lated Cerasome was modified with chelating agent DOTA for 111 In-labeling. Normal mice firstly were used for in vivo studies. Animals were sacrificed at different time points after tail vein administration, blood samples were taken and the organs of interest were captured to evaluate the pharmacokinetic properties and in vivo biodistribution of 111 In-ICG-DPDCs. The subcutaneous Lewis lung carcinoma ( LLC ) tumor model in C57BL/6 mouse was established. The tumor-bearing mice were subjected to optical imaging in small animal IVIS and SPECT imaging in small animal nanoScanSPECT/CT system for tumor uptake of 111 In-ICG-DPDCs. Results The size of the nanoparticle probe was about 90 nm, and the 111 In-labeling was successfully per-formed with 99.93% radiochemical purity after purification. 111 In-ICG-DPDCs showed excellent in vitro sta-bility with 97.10% radiochemical purity at 48 h post-purification. In vivo blood clearance experiments showed that 111 In-ICG-DPDCs had a relative long blood circulation time with the fast and slow phase half-lives of 40 and 132.7 min. 111In-ICG-DPDCs accumulated mainly in the liver and spleen, with long retention time. NanoScanSPECT/CT imaging showed that LLC tumors were significantly visualized at 4 h post-injection, and the other major accumulated organs were the liver and spleen, which were consistent with the results of biodistribution. Optical imaging showed significant uptake of the nanoparticle probe in the tumor, confirming the SPECT imaging results. Conclusion The Cerasome based probe designed could be used for tumor SPECT and optical dual-modality imaging, and has potential for therapeutic use.

A novel integrinαvβ3-targeted isoDGR probe for SPECT/CT imaging of glioma / 中华核医学与分子影像杂志

Objective To prepare 99 Tcm-HYNIC-c( isoDGRKy) as a SPECT/CT imaging molecu-lar probe targeting integrin αvβ3 , and evaluate its biodistribution and feasibility on SPECT/CT imaging for integrinαvβ3-positive tumor in U87MG human glioma xenograft mouse models. Methods The bifunctional chelator HYNIC was conjugated to c( isoDGRKy) , and tricine and TPPTS were used as coligands for 99 Tcm labeling to prepare 99 Tcm-HYNIC-c( isoDGRKy) . The radiochemical purity and stability of the product were measured. The expression of integrin αvβ3 and binding affinity ( half maximal inhibitory concentration, IC50 ) of c ( isoDGRKy ) was detected in U87MG cells by cell experiments in vitro. Biodistribution and SPECT/CT imaging of 99 Tcm-HYNIC-c( isoDGRKy) including blocking experiments were performed respec-tively in nude mice bearing U87MG human glioma xenografts. Results The radiochemical purity of 99 Tcm-HYNIC-c( isoDGRKy) was over 99%, and was still over 99% after 4 h incubation in saline at room temper-ature. Flow cytometry assay showed that U87MG cells were integrinαvβ3-positive ( expressive rate:70%) . The IC50 of c(isoDGRKy) was 6.67×10-8 mol/L. Biodistribution results showed 99Tcm-HYNIC-c(isoDGRKy) with a rapid clearance from blood was excreted mainly via the kidneys. The 99 Tcm-HYNIC-c( isoDGRKy) uptake values in U87MG tumors were (7.31±1.42) and (1.09±0.11) %ID/g at 15 and 45 min post-injection re-spectively, and tumor-to-muscle ratio reached 5.01±1.47 at 15 min post-injection. The tumors were clearlyvisualized with low background from 0.5 to 1 h post-injection in tumor bearing mice. In the blocking experi-ment, the tumor was barely visualized after co-injection of excess cold c(RGDfK) peptide with 99Tcm-HYNIC-c(isoDGRKy). Conclusions 99Tcm-HYNIC-c(isoDGRKy) may be easily and steadily prepared. It may be a RGD-like promising SPECT/CT imaging probe for integrinαvβ3-positive tumor.

Synthesis and in vivo evaluation of 68Ga-DOTATATE in mice / 中华核医学与分子影像杂志

Objective To synthesize 68 Ga-1,4,7,10-tetraazacyclododecane-N,N',N,N()-tetraacetic acid-D-Phe1-Tyr3-Thr8-octreotide (68Ga-DOTATATE) manually and automatically,validate its qualities in vitro,and evaluate its biodistribution in ICR mice and the microPET imaging in nude mice bearing pancreatic AR42J tumor.Methods 68Ga-DOTATATE was synthesized by automatic method using commercial metal isotope multifunctional module with strong cation exchange (SCX) column and by manual method.Both the products were measured for quality control.For the biodistribution study 5 groups of ICR mice were injected with 68Ga-DOTATATE(1.11 MBq) and executed at 10,30,60,120 and 240 min postinjection,respectively.The organs were weighted,and ％ ID/g was calculated.Nude mice bearing pancreatic AR42J tumor were intravenously injected with 3.7 MBq 68Ga-DOTATATE,and then microPET imaging was acquired at 10,30,60,120,18 and 240 min.Results 68Ga-DOTATATE could be successfully synthesized by the automatic and manual methods.Both the product injections were colorless and clear.The pH value was 6.5±0.1.For the products obtained from the two methods,the radiochemical purities were over 99％,and the products were stable for 4 h at room temperature.For the automatic method,68Ga-DOTATATE was synthesized within 30 min and with the radiochemical yield of (51.8±3.2)％ (time decay corrected).For the manual method,the time used for the synthesis was 20 min,and the labeling yield was over 99％.Three batches of the products were aseptic and pyrogen-free.In ICR mice,68Ga-DOTATATE was excreted by the kidney,and showed relatively high accumulation in the liver,spleen,pancreas and adrenal glands,while lower in the bone and soft tissue.The clearance from blood was fast with (4.41±0.81) ％ID/g at 10 min postinjection and (0.78 ± 0.32) ％ ID/g at 1 h.MicroPET imaging showed increased uptake of 68GaDOTATATE in the tumor tissues,and T/NT were 2.01±0.29(10 min),6.74±2.90(30 min) and 4.46±2.05 (60 min),respectively.Conclusions 68 Ga-DOTATATE could be successfully synthesized manually and automatically.The products reach to the specification of radioactive drugs and could be used as an attractive positron emitting radiotracer for detection of the somatostatin receptor-positive tumors.