Intravoxel incoherent motion radiomics nomogram for predicting tumor treatment responses in nasopharyngeal carcinoma

BackgroundIntravoxel incoherent motion (IVIM) plays an important role in predicting treatment responses in patient with nasopharyngeal carcinoma (NPC). The goal of this study was to develop and validate a radiomics nomogram based on IVIM parametric maps and clinical data for the prediction of treatment responses in NPC patients.MethodsEighty patients with biopsy-proven NPC were enrolled in this study. Sixty-two patients had complete responses and 18 patients had incomplete responses to treatment. Each patient received a multiple b-value diffusion-weighted imaging (DWI) examination before treatment. Radiomics features were extracted from IVIM parametric maps derived from DWI image. Feature selection was performed by the least absolute shrinkage and selection operator method. Radiomics signature was generated by support vector machine based on the selected features. Receiver operating characteristic (ROC) curves and area under the ROC curve (AUC) values were used to evaluate the diagnostic performance of radiomics signature. A radiomics nomogram was established by integrating the radiomics signature and clinical data.ResultsThe radiomics signature showed good prognostic performance to predict treatment response in both training (AUC = 0.906, P<0.001) and testing (AUC = 0.850, P<0.001) cohorts. The radiomic nomogram established by integrating the radiomic signature with clinical data significantly outperformed clinical data alone (C-index, 0.929 vs 0.724; P<0.0001).ConclusionsThe IVIM-based radiomics nomogram provided high prognostic ability to treatment responses in patients with NPC. The IVIM-based radiomics signature has the potential to be a new biomarker in prediction of the treatment responses and may affect treatment strategies in patients with NPC.     

Application of radiography of computed tomography in non-small cell lung cancer using prognosis model

ObjectiveStudying the diagnostic value of CT imaging in non-small cell lung cancer (NSCLC), and establishing a prognosis model combined with clinical characteristics is the objective, so as to provide a reference for the survival prediction of NSCLC patients.MethodCT scan data of NSCLC 200 patients were taken as the research object. Through image segmentation, the radiology features of CT images were extracted. The reliability and performance of the prognosis model based on the optimal feature number of specific algorithm and the prognosis model based on the global optimal feature number were compared.Results30-RELF-NB (30 optimal features, RELF feature selection algorithm and NB classifier) has the highest accuracy and AUC (area under the subject characteristic curve) in the prognosis model based on the optimal features of specific algorithm. Among the prognosis models based on global optimal features, 25-NB (25 global optimal features, naive Bayes classification algorithm classifier) has the highest accuracy and AUC. Compared with the prediction model based on feature training of specific feature selection algorithm, the overall performance and stability of the prediction model based on global optimal feature are higher.ConclusionThe prognosis model based on the global optimal feature established in this paper has good reliability and performance, and can be applied to the CT radiology of NSCLC.     

Computed tomography-based radiomic model at node level for the prediction of normal-sized lymph node metastasis in cervical cancer

PurposeRadiomic models have been demonstrated to have acceptable discrimination capability for detecting lymph node metastasis (LNM). We aimed to develop a computed tomography–based radiomic model and validate its usefulness in the prediction of normal-sized LNM at node level in cervical cancer.MethodsA total of 273 LNs of 219 patients from 10 centers were evaluated in this study. We randomly divided the LNs from the 2 centers with the largest number of LNs into the training and internal validation cohorts, and the rest as the external validation cohort. Radiomic features were extracted from the arterial and venous phase images. We trained an artificial neural network (ANN) to develop two single-phase models. A radiomic model reflecting the features of two-phase images was also built for directly predicting LNM in cervical cancer. Moreover, four state-of-the-art methods were used for comparison. The performance of all models was assessed using the area under the receiver operating characteristic curve (AUC).ResultsAmong the models we built, the models combining the features of two phases surpassed the single-phase models, and the models generated by ANN had better performance than the others. We found that the radiomic model achieved the highest AUCs of 0.912 and 0.859 in the training and internal validation cohorts, respectively. In the external validation cohort, the AUC of the radiomic model was 0.800.ConclusionWe constructed a radiomic model that exhibited great ability in the prediction of LNM. The application of the model could optimize clinical staging and decision-making.     

Deep learning dose prediction for IMRT of esophageal cancer: The effect of data quality and quantity on model performance

PurposeTo investigate the effect of data quality and quantity on the performance of deep learning (DL) models, for dose prediction of intensity-modulated radiotherapy (IMRT) of esophageal cancer.Material and methodsTwo databases were used: a variable database (VarDB) with 56 clinical cases extracted retrospectively, including user-dependent variability in delineation and planning, different machines and beam configurations; and a homogenized database (HomDB), created to reduce this variability by re-contouring and re-planning all patients with a fixed class-solution protocol.Experiment 1 analysed the user-dependent variability, using 26 patients planned with the same machine and beam setup (E26-VarDB versus E26-HomDB). Experiment 2 increased the training set by groups of 10 patients (E16, E26, E36, E46, and E56) for both databases.Model evaluation metrics were the mean absolute error (MAE) for selected dose-volume metrics and the global MAE for all body voxels.ResultsFor Experiment 1, E26-HomDB reduced the MAE for the considered dose-volume metrics compared to E26-VarDB (e.g. reduction of 0.2 Gy for D95-PTV, 1.2 Gy for Dmean-heart or 3.3% for V5-lungs). For Experiment 2, increasing the database size slightly improved performance for HomDB models (e.g. decrease in global MAE of 0.13 Gy for E56-HomDB versus E26-HomDB), but increased the error for the VarDB models (e.g. increase in global MAE of 0.20 Gy for E56-VarDB versus E26-VarDB).ConclusionA small database may suffice to obtain good DL prediction performance, provided that homogenous training data is used. Data variability reduces the performance of DL models, which is further pronounced when increasing the training set.     

Normal tissue complication probability models for severe acute radiological lung injury after radiotherapy for lung cancer

PurposeTo derive Normal Tissue Complication Probability (NTCP) models for severe patterns of early radiological radiation-induced lung injury (RRLI) in patients treated with radiotherapy (RT) for lung tumors. Second, derive threshold doses and optimal doses for prediction of RRLI to be used in differential diagnosis of tumor recurrence from RRLI during follow-up.Methods and materialsLyman-EUD (LEUD), Logit-EUD (LogEUD), relative seriality (RS) and critical volume (CV) NTCP models, with DVH corrected for fraction size, were used to model the presence of severe early RRLI in follow-up CTs. The models parameters, including α/β, were determined by fitting data from forty-five patients treated with IMRT for lung cancer. Models were assessed using Akaike information criterion (AIC) and area under receiver operating characteristic curve (AUC). Threshold doses for risk of RRLI and doses corresponding to the optimal point of the receiver operating characteristic (ROC) curve were determined.ResultsThe α/βs obtained with different models were 2.7–3.2 Gy. The thresholds and optimal doses curves were EUDs of 3.2–7.8 Gy and 15.2–18.1 Gy with LEUD, LogEUD and RS models, and μ_d of 0.013 and 0.071 with the CV model. NTCP models had AUCs significantly higher than 0.5. Occurrence and severity of RRLI were correlated with patients’ values of EUD and μ_d.ConclusionsThe models and dose levels derived can be used in differential diagnosis of tumor recurrence from RRLI in patients treated with RT. Cross validation is needed to prove prediction performance of the model outside the dataset from which it was derived.     

Early risk-assessment of patients with nasopharyngeal carcinoma: the added prognostic value of MR-based radiomics

ObjectivesTo assess the additive prognostic value of MR-based radiomics in predicting progression-free survival (PFS) in patients with nasopharyngeal carcinoma (NPC)MethodsPatients newly diagnosed with non-metastatic NPC between June 2006 and October 2019 were retrospectively included and randomly grouped into training and test cohorts (7:3 ratio). Radiomic features (n=213) were extracted from T2-weighted and contrast-enhanced T1-weighted MRI. The patients were staged according to the 8^th edition of American Joint Committee on Cancer Staging Manual. The least absolute shrinkage and selection operator was used to select the relevant radiomic features. Univariate and multivariate Cox proportional hazards analyses were conducted for PFS, yielding three different survival models (clinical, stage, and radiomic). The integrated time-dependent area under the curve (iAUC) for PFS was calculated and compared among different combinations of survival models, and the analysis of variance was used to compare the survival models. The prognostic performance of all models was validated using a test set with integrated Brier scores.ResultsThis study included 81 patients (training cohort=57; test cohort=24), and the mean PFS was 57.5 ± 43.6 months. In the training cohort, the prognostic performances of survival models improved significantly with the addition of radiomics to the clinical (iAUC, 0.72–0.80; p=0.04), stage (iAUC, 0.70–0.79; p=0.001), and combined models (iAUC, 0.76–0.81; p<0.001). In the test cohort, the radiomics and combined survival models were robustly validated for their ability to predict PFS.ConclusionIntegration of MR-based radiomic features with clinical and stage variables improved the prediction PFS in patients diagnosed with NPC.     

CT radiomics combined with clinical variables for predicting the overall survival of hepatocellular carcinoma patients after hepatectomy

PurposeTo establish a model for assessing the overall survival (OS) of the hepatocellular carcinoma (HCC) patients after hepatectomy based on the clinical and radiomics features.MethodsThis study recruited a total of 267 patients with HCC, which were randomly divided into the training (N = 188) and validation (N = 79) cohorts. In the training cohort, radiomic features were selected with the intra-reader and inter-reader correlation coefficient (ICC), Spearman's correlation coefficient, and the least absolute shrinkage and selection operator (LASSO). The radiomics signatures were built by COX regression analysis and compared the predictive potential in the different phases (arterial, portal, and double-phase) and regions of interest (tumor, peritumor 3 mm, peritumor 5 mm). A clinical-radiomics model (CR model) was established by combining the radiomics signatures and clinical risk factors. The validation cohort was used to validate the proposed models.ResultsA total of 267 patients 86 (45.74%) and 37 (46.84%) patients died in the training and validation cohorts, respectively. Among all the radiomics signatures, those based on the tumor and peritumor (5 mm) (AP-TP5-Signature) showed the best prognostic potential (training cohort 1–3 years AUC:0.774–0.837; validation cohort 1–3 years AUC:0.754–0.810). The CR model showed better discrimination, calibration, and clinical applicability as compared to the clinical model and radiomics features. In addition, the CR model could perform risk-stratification and also allowed for significant discrimination between the Kaplan-Meier curves in most of the subgroups.ConclusionsThe CR model could predict the OS of the HCC patients after hepatectomy.     

Deep radiomic model based on the sphere–shell partition for predicting treatment response to chemotherapy in lung cancer

BackgroundThe prognosis of chemotherapy is important in clinical decision-making for non-small cell lung cancer (NSCLC) patients.ObjectivesTo develop a model for predicting treatment response to chemotherapy in NSCLC patients from pre-chemotherapy CT images.Materials and MethodsThis retrospective multicenter study enrolled 485 patients with NSCLC who received chemotherapy alone as a first-line treatment. Two integrated models were developed using radiomic and deep-learning-based features. First, we partitioned pre-chemotherapy CT images into spheres and shells with different radii around the tumor (0–3, 3–6, 6–9, 9–12, 12–15 mm) containing intratumoral and peritumoral regions. Second, we extracted radiomic and deep-learning-based features from each partition. Third, using radiomic features, five sphere–shell models, one feature fusion model, and one image fusion model were developed. Finally, the model with the best performance was validated in two cohorts.ResultsAmong the five partitions, the model of 9–12 mm achieved the highest area under the curve (AUC) of 0.87 (95% confidence interval: 0.77–0.94). The AUC was 0.94 (0.85–0.98) for the feature fusion model and 0.91 (0.82–0.97) for the image fusion model. For the model integrating radiomic and deep-learning-based features, the AUC was 0.96 (0.88–0.99) for the feature fusion method and 0.94 (0.85–0.98) for the image fusion method. The best-performing model had an AUC of 0.91 (0.81–0.97) and 0.89 (0.79–0.93) in two validation sets, respectively.ConclusionsThis integrated model can predict the response to chemotherapy in NSCLC patients and assist physicians in clinical decision-making.     

The amputation and survival of patients with diabetic foot based on establishment of prediction model

ObjectiveThe objective of this paper is to study the establishment of predictive models and the amputation and survival of patients with diabetic foot.MethodsA total of 200 inpatients with diabetic foot were selected as the research subject in this study. The amputation and survival status of diabetic foot patients were followed up by telephone. The relevant indicators were screened by cluster analysis. The predictive model was established respectively based on proportional hazard regression analysis, back propagation neural network (BPNN) and BPNN based on genetic algorithm optimization, and the reliability of the three prediction models (PM) was evaluated and compared.ResultsThe risk factors for amputation were severe ulcer disease, glycosylated hemoglobin and low-density lipoprotein cholesterol. The risk factors for death were cerebrovascular disease, severe ulcer disease and peripheral arterial disease. In case that the outcome was amputation, the PM of BPNN and the PM of BPNN based on genetic algorithm optimization have obviously higher AUC (area under the receiver operating characteristic curve) than the PM of proportional hazard regression analysis, and the difference was statistically significant (P < 0.05). Among the three PMs, the PM based on BPNN had the highest AUC, sensitivity and specificity (SAS). In case that the outcome was death, the PM of BPNN and the PM of BPNN based on genetic algorithm optimization had almost the same AUC, and were obviously higher than the PM based on proportional hazard regression analysis. The difference was statistically significant (P < 0.05). The PM based on BPNN and the BPNN based on genetic algorithm optimization had higher SAS than the PM based on COX regression analysis.ConclusionThe PM of BPNN and BPNN based on genetic algorithm optimization have better prediction effect than the PM based on proportional hazard regression analysis. It can be used for amputation and survival analysis of diabetic foot patients.     

Accurate Prediction of Transposon-Derived piRNAs by Integrating Various Sequential and Physicochemical Features

BackgroundPiwi-interacting RNA (piRNA) is the largest class of small non-coding RNA molecules. The transposon-derived piRNA prediction can enrich the research contents of small ncRNAs as well as help to further understand generation mechanism of gamete.MethodsIn this paper, we attempt to differentiate transposon-derived piRNAs from non-piRNAs based on their sequential and physicochemical features by using machine learning methods. We explore six sequence-derived features, i.e. spectrum profile, mismatch profile, subsequence profile, position-specific scoring matrix, pseudo dinucleotide composition and local structure-sequence triplet elements, and systematically evaluate their performances for transposon-derived piRNA prediction. Finally, we consider two approaches: direct combination and ensemble learning to integrate useful features and achieve high-accuracy prediction models.ResultsWe construct three datasets, covering three species: Human, Mouse and Drosophila, and evaluate the performances of prediction models by 10-fold cross validation. In the computational experiments, direct combination models achieve AUC of 0.917, 0.922 and 0.992 on Human, Mouse and Drosophila, respectively; ensemble learning models achieve AUC of 0.922, 0.926 and 0.994 on the three datasets.ConclusionsCompared with other state-of-the-art methods, our methods can lead to better performances. In conclusion, the proposed methods are promising for the transposon-derived piRNA prediction. The source codes and datasets are available in S1 File.     

Machine learning-based in-hospital mortality prediction of HIV/AIDS patients with Talaromyces marneffei infection in Guangxi,China

ObjectiveTalaromycosis is a serious regional disease endemic in Southeast Asia. In China, Talaromyces marneffei (T. marneffei) infections is mainly concentrated in the southern region, especially in Guangxi, and cause considerable in-hospital mortality in HIV-infected individuals. Currently, the factors that influence in-hospital death of HIV/AIDS patients with T. marneffei infection are not completely clear. Existing machine learning techniques can be used to develop a predictive model to identify relevant prognostic factors to predict death and appears to be essential to reducing in-hospital mortality.MethodsWe prospectively enrolled HIV/AIDS patients with talaromycosis in the Fourth People’s Hospital of Nanning, Guangxi, from January 2012 to June 2019. Clinical features were selected and used to train four different machine learning models (logistic regression, XGBoost, KNN, and SVM) to predict the treatment outcome of hospitalized patients, and 30% internal validation was used to evaluate the performance of models. Machine learning model performance was assessed according to a range of learning metrics, including area under the receiver operating characteristic curve (AUC). The SHapley Additive exPlanations (SHAP) tool was used to explain the model.ResultsA total of 1927 HIV/AIDS patients with T. marneffei infection were included. The average in-hospital mortality rate was 13.3% (256/1927) from 2012 to 2019. The most common complications/coinfections were pneumonia (68.9%), followed by oral candida (47.5%), and tuberculosis (40.6%). Deceased patients showed higher CD4/CD8 ratios, aspartate aminotransferase (AST) levels, creatinine levels, urea levels, uric acid (UA) levels, lactate dehydrogenase (LDH) levels, total bilirubin levels, creatine kinase levels, white blood-cell counts (WBC) counts, neutrophil counts, procaicltonin levels and C-reactive protein (CRP) levels and lower CD3⁺ T-cell count, CD8⁺ T-cell count, and lymphocyte counts, platelet (PLT), high-density lipoprotein cholesterol (HDL), hemoglobin (Hb) levels than those of surviving patients. The predictive XGBoost model exhibited 0.71 sensitivity, 0.99 specificity, and 0.97 AUC in the training dataset, and our outcome prediction model provided robust discrimination in the testing dataset, showing an AUC of 0.90 with 0.69 sensitivity and 0.96 specificity. The other three models were ruled out due to poor performance. Septic shock and respiratory failure were the most important predictive features, followed by uric acid, urea, platelets, and the AST/ALT ratios.ConclusionThe XGBoost machine learning model is a good predictor in the hospitalization outcome of HIV/AIDS patients with T. marneffei infection. The model may have potential application in mortality prediction and high-risk factor identification in the talaromycosis population.     

Multiparametric MRI-based radiomics analysis for the prediction of breast tumor regression patterns after neoadjuvant chemotherapy

ObjectivesBreast cancers show different regression patterns after neoadjuvant chemotherapy. Certain regression patterns are associated with more reliable margins in breast-conserving surgery. Our study aims to establish a nomogram based on radiomic features and clinicopathological factors to predict regression patterns in breast cancer patients.MethodsWe retrospectively reviewed 144 breast cancer patients who received neoadjuvant chemotherapy and underwent definitive surgery in our center from January 2016 to December 2019. Tumor regression patterns were categorized as type 1 (concentric regression + pCR) and type 2 (multifocal residues + SD + PD) based on pathological results. We extracted 1158 multidimensional features from 2 sequences of MRI images. After feature selection, machine learning was applied to construct a radiomic signature. Clinical characteristics were selected by backward stepwise selection. The combined prediction model was built based on both the radiomic signature and clinical factors. The predictive performance of the combined prediction model was evaluated.ResultsTwo radiomic features were selected for constructing the radiomic signature. Combined with two significant clinical characteristics, the combined prediction model showed excellent prediction performance, with an area under the receiver operating characteristic curve of 0.902 (95% confidence interval 0.8343–0.9701) in the primary cohort and 0.826 (95% confidence interval 0.6774–0.9753) in the validation cohort.ConclusionsOur study established a unique model combining a radiomic signature and clinicopathological factors to predict tumor regression patterns prior to the initiation of NAC. The early prediction of type 2 regression offers the opportunity to modify preoperative treatments or aids in determining surgical options.     

Local control and failure patterns after intensity modulated radiotherapy with reduced target volume delineation after induction chemotherapy for patients with T4 nasopharyngeal carcinoma

BackgroundThe delineation of target volume after induction chemotherapy(IC) for nasopharyngeal carcinoma(NPC) is currently controversial. In this study, we aimed to analyze the long-term local control(LC) and failure patterns of T4 NPC treated with reduced target volume radiotherapy after IC.MethodsFrom September 2007 to January 2013, 145 patients with T4 NPC were retrospectively reviewed. All patients received at least 1 cycle of IC followed by intensity modulated radiotherapy(IMRT). The gross tumor volume(GTV) was delineated according to the post-IC images for intracavity tumors and lymph nodes. The LC and overall survival (OS) rates were calculated using the Kaplan-Meier method. The location and extent of local failures were transferred to the pretreatment planning computed tomography (CT) for dosimetric analysis.ResultsWith a median follow-up time of 95 months (range, 16–142 months), 23 local failures were found. The estimated 10-year LC and OS rates were 81.1%and 54.8% respectively. Among the 20 local failures with available diagnostic images, 18(90%) occurred within the 95% isodose lines and were considered in-field failures and 2(10%) were marginal. There was no outside-field failure.ConclusionsIn-field failure was the major pattern of local failure for T4 NPC. IMRT with reduced target volume after IC seems to be feasible. Further researches exploring optimal volume and radiation dose for local advanced NPC in the era of IC are warranted.     

Integral dose based inverse optimization objective function promises lower toxicity in head-and-neck

PurposeThe voxels in a CT data sets contain density information. Besides its use in dose calculation density has no other application in modern radiotherapy treatment planning. This work introduces the use of density information by integral dose minimization in radiotherapy treatment planning for head-and-neck squamous cell carcinoma (HNSCC).Materials and methodsEighteen HNSCC cases were studied. For each case two intensity modulated radiotherapy (IMRT) plans were created: one based on dose-volume (DV) optimization, and one based on integral dose minimization (Energy hereafter) inverse optimization. The target objective functions in both optimization schemes were specified in terms of minimum, maximum, and uniform doses, while the organs at risk (OAR) objectives were specified in terms of DV- and Energy-objectives respectively. Commonly used dosimetric measures were applied to assess the performance of Energy-based optimization. In addition, generalized equivalent uniform doses (gEUDs) were evaluated. Statistical analyses were performed to estimate the performance of this novel inverse optimization paradigm.ResultsEnergy-based inverse optimization resulted in lower OAR doses for equivalent target doses and isodose coverage. The statistical tests showed dose reduction to the OARs with Energy-based optimization ranging from ∼2% to ∼15%.ConclusionsIntegral dose minimization based inverse optimization for HNSCC promises lower doses to nearby OARs. For comparable therapeutic effect the incorporation of density information into the optimization cost function allows reduction in the normal tissue doses and possibly in the risk and the severity of treatment related toxicities.     

Clinicopathologic predictors of early relapse in advanced epithelial ovarian cancer: development of prediction models using nationwide data

ObjectiveTo identify clinicopathologic factors predictive of early relapse (platinum-free interval (PFI) of ≤6 months) in advanced epithelial ovarian cancer (EOC) in first-line treatment, and to develop and internally validate risk prediction models for early relapse.MethodsAll consecutive patients diagnosed with advanced stage EOC between 01-01-2008 and 31-12-2015 were identified from the Netherlands Cancer Registry. Patients who underwent cytoreductive surgery and platinum-based chemotherapy as initial EOC treatment were selected. Two prediction models, i.e. pretreatment and postoperative, were developed. Candidate predictors of early relapse were fitted into multivariable logistic regression models. Model performance was assessed on calibration and discrimination. Internal validation was performed through bootstrapping to correct for model optimism.ResultsA total of 4,557 advanced EOC patients were identified, including 1,302 early relapsers and 3,171 late or non-relapsers. Early relapsers were more likely to have FIGO stage IV, mucinous or clear cell type EOC, ascites, >1 cm residual disease, and to have undergone NACT-ICS. The final pretreatment model demonstrated subpar model performance (AUC = 0.64 [95 %-CI 0.62−0.66]). The final postoperative model based on age, FIGO stage, pretreatment CA-125 level, histologic subtype, presence of ascites, treatment approach, and residual disease after debulking, demonstrated adequate model performance (AUC = 0.72 [95 %-CI 0.71−0.74]). Bootstrap validation revealed minimal optimism of the final postoperative model.ConclusionA (postoperative) discriminative model has been developed and presented online that predicts the risk of early relapse in advanced EOC patients. Although external validation is still required, this prediction model can support patient counselling in daily clinical practice.     

Delta radiomics for rectal cancer response prediction using low field magnetic resonance guided radiotherapy: an external validation

IntroductionA recent study performed on 16 locally advanced rectal cancer (LARC) patients treated using magnetic resonance guided radiotherapy (MRgRT) has identified two delta radiomics features as predictors of clinical complete response (cCR) after neoadjuvant radio-chemotherapy (nCRT).This study aims to validate these features (ΔL_least and Δglnu) on an external larger dataset, expanding the analysis also for pathological complete response (pCR) prediction.MethodsA total of 43 LARC patients were enrolled: Gross Tumour Volume (GTV) was delineated on T2/T1* MR images acquired during MRgRT and the two delta features were calculated.Receiver Operating Characteristic (ROC) curve analysis was performed on the 16 cases of the original study and the best cut-off value was identified. The performance of ΔL_least and Δglnu was evaluated at the best cut-off value.ResultsOn the original dataset of 16 patients, ΔL_least reported an AUC of 0.81 for cCR and 0.93 for pCR, while Δglnu 0.72 and 0.54 respectively.The best cut-off values of ΔL_least was 0.73 for both outcomes, while Δglnu reported 0.54 for cCR and 0.93 for pCR. At the external validation, ΔL_least showed an accuracy of 81% for cCR and 79% for pCR, while Δglnu reported 63% for cCR and 40% for pCR.ConclusionThe accuracy of ΔL_least in predicting cCR and pCR is significantly higher than those obtained considering Δglnu, but inferior if compared with other image-based biomarker, such as the early-regression index. Studies with larger cohorts of patients are recommended to further investigate the role of delta radiomic features in MRgRT.     

Which evaluation criteria of the short-term efficacy can better reflect the long-term outcomes for patients with nasopharyngeal carcinoma?

PurposeTo compare the consistency of one-dimensional Response Evaluation Criteria in Solid Tumors (1D-RECIST), two-dimensional WHO criteria (2D-WHO), and three-dimensional (3D) measurement for therapeutic response assessment of nasopharyngeal carcinoma (NPC).Materials and methodsRetrospective data of 288 newly diagnosed NPC patients were reviewed. Tumor size was assessed on magnetic resonance imaging (MRI) according to the 1D-RECIST, 2D-WHO, and 3D measurement criteria. Agreement between tumor responses was assessed using unweighted k statistics. The receiver operating characteristic (ROC) curve was used to determine the optimal cut-off point of the PTV. The Kaplan–Meier method and Cox regression were used for the survival analysis.ResultsThe optimal cut-off point of the PTV for progression-free survival (PFS) was 29.6%. Agreement with PTV measurement was better for 1D measurement than for 2D and 3D measurements (kappa values of 0.646, 0.537, and 0.577 for 1D, 2D, and 3D measurements, respectively; P < 0.05). The area under the curve of the 1D measurement (AUC=0.596) was similar to that of the PTV measurement (AUC=0.621). Compared with 2D and 3D measurements, 1D measurement is superior for predicting prognosis in NPC (C-index of 0.672, 0.663, and 0.646 were for 1D, 2D, and 3D measurements, respectively; P < 0.005). Survival analysis showed that patients with non-responders had worse prognosis (P < 0.05).ConclusionsThe 1D measurement more closely agreed with the PTV measurement than the 2D and 3D measurements for predicting therapeutic responses in NPC. Therefore, we recommend using the less time-consuming 1D-RECIST criteria in routine clinical practice.     

Prognosis Analysis of Heart Failure Based on Recurrent Attention Model

ObjectivesHeart failure is a group of complex clinical syndromes that lead to ventricular filling or impaired ejection ability due to abnormal heart structure or function. Difficult treatment, poor prognosis and high mortality are the main characteristics of heart failure. According to admission data and past medical use, the 30-day mortality rate of patients with heart failure was obtained and the main characteristics affecting the 30-day mortality of patients with heart failure were determined.Material and methodsBased on the data of April 2016 to July 2018 of Shanxi Acadeny of Medical Sciences, and we chose 4,682 information on heart failure patients, of which 539 died in the hospital by screening. We built a 30-day mortality prediction model for patients with heart failure. The model can fuse clinical data and text data through multiple kernel learning, and input the fused data into the recurrent attention model. It can not only predict the 30-day mortality of patients with heart failure, but also the influencing factors of prognosis of patients with heart failure were also obtained.ResultsThe prediction accuracy of the recurrent attention network is obviously higher than that of other machine learning models, and the accuracy rate reaches 93.4%. The AUC value of the area under the ROC curve of the model reaches 87%, which is obviously higher than that of the traditional machine learning models such as decision tree, naive Bayesian and support vector machine. In addition, the model can also reach a conclusion that New York heart function classification, age, NT—ProBNP, LVEF, β-blockers, ventricular arrhythmia, high blood pressure, coronary heart disease (CHD) and bronchitis were independent risk factors for death. And patients with revascularization, ACEI/ARB drugs, β-blockers, spironolactone have a better prognosis than non-users. This provides an important reference for doctors to better treat and manage patients with heart failure.ConclusionExperiments show that the prognostic effect of the recurrent attention model is significantly higher than that of other traditional machine learning models. Because the model increases the attention mechanism, the important features affecting the prognostic results are obtained, which enables doctors to prescribe drugs according to the symptoms, take timely precautions and help patients to treat in time.     

Bayesian approach for the estimation of cyclosporine area under the curve using limited sampling strategies in pediatric hematopoietic stem cell transplantation

Background

The optimal marker for cyclosporine (CsA) monitoring in transplantation patients remains controversial. However, there is a growing interest in the use of the area under the concentration-time curve (AUC), particularly for cyclosporine dose adjustment in pediatric hematopoietic stem cell transplantation. In this paper, we develop Bayesian limited sampling strategies (B-LSS) for cyclosporine AUC estimation using population pharmacokinetic (Pop-PK) models and investigate related issues, with the aim to improve B-LSS prediction performance.

Methods

Twenty five pediatric hematopoietic stem cell transplantation patients receiving intravenous and oral cyclosporine were investigated. Pop-PK analyses were carried out and the predictive performance of B-LSS was evaluated using the final Pop-PK model and several related ones. The performance of B-LSS when targeting different versions of AUC was also discussed.

Results

A two-compartment structure model with a lag time and a combined additive and proportional error is retained. The final covariate model does not improve the B-LSS prediction performance. The best performing models for intravenous and oral cyclosporine are the structure ones with combined and additive error, respectively. Twelve B-LSS, consisting of 4 or less sampling points obtained within 4 hours post-dose, predict AUC with 95^th percentile of the absolute values of relative prediction errors of 20% or less. Moreover, B-LSS perform better for the prediction of the ‘underlying’ AUC derived from the Pop-PK model estimated concentrations that exclude the residual errors, in comparison to their prediction of the observed AUC directly calculated using measured concentrations.

Conclusions

B-LSS can adequately estimate cyclosporine AUC. However, B-LSS performance is not perfectly in line with the standard Pop-PK model selection criteria; hence the final model might not be ideal for AUC prediction purpose. Therefore, for B-LSS application, Pop-PK model diagnostic criteria should additionally account for AUC prediction errors.