test sets
Recently Published Documents





2022 ◽  
Vol 11 ◽  
Haolin Yin ◽  
Yu Jiang ◽  
Zihan Xu ◽  
Wenjun Huang ◽  
Tianwu Chen ◽  

Background and PurposeBreast ductal carcinoma in situ (DCIS) has no metastatic potential, and has better clinical outcomes compared with invasive breast cancer (IBC). Convolutional neural networks (CNNs) can adaptively extract features and may achieve higher efficiency in apparent diffusion coefficient (ADC)-based tumor invasion assessment. This study aimed to determine the feasibility of constructing an ADC-based CNN model to discriminate DCIS from IBC.MethodsThe study retrospectively enrolled 700 patients with primary breast cancer between March 2006 and June 2019 from our hospital, and randomly selected 560 patients as the training and validation sets (ratio of 3 to 1), and 140 patients as the internal test set. An independent external test set of 102 patients during July 2019 and May 2021 from a different scanner of our hospital was selected as the primary cohort using the same criteria. In each set, the status of tumor invasion was confirmed by pathologic examination. The CNN model was constructed to discriminate DCIS from IBC using the training and validation sets. The CNN model was evaluated using the internal and external tests, and compared with the discriminating performance using the mean ADC. The area under the curve (AUC), sensitivity, specificity, and accuracy were calculated to evaluate the performance of the previous model.ResultsThe AUCs of the ADC-based CNN model using the internal and external test sets were larger than those of the mean ADC (AUC: 0.977 vs. 0.866, P = 0.001; and 0.926 vs. 0.845, P = 0.096, respectively). Regarding the internal test set and external test set, the ADC-based CNN model yielded sensitivities of 0.893 and 0.873, specificities of 0.929 and 0.894, and accuracies of 0.907 and 0.902, respectively. Regarding the two test sets, the mean ADC showed sensitivities of 0.845 and 0.818, specificities of 0.821 and 0.829, and accuracies of 0.836 and 0.824, respectively. Using the ADC-based CNN model, the prediction only takes approximately one second for a single lesion.ConclusionThe ADC-based CNN model can improve the differentiation of IBC from DCIS with higher accuracy and less time.

2022 ◽  
Vol 9 ◽  
Magdalena Postek ◽  
Katarzyna Walicka-Serzysko ◽  
Justyna Milczewska ◽  
Dorota Sands

IntroductionIn cystic fibrosis (CF), pathological lung changes begin early in life. The technological progress currently gives many diagnostic possibilities. However, pulmonary function testing in children remains problematic.ObjectivesOur study aimed to correlate the results of impulse oscillometry (IOS) with those of multiple breath nitrogen washout (MBNW) in our pediatric CF population. We also compared those parameters between the groups with and without spirometric features of obturation.MethodsWe collected 150 pulmonary function test sets, including spirometry, IOS, and MBNW in patients with CF aged 12.08 ± 3.85 years [6–18]. The study group was divided into two subgroups: IA (without obturation) and IB (with obturation). We also compared Sacin, Scond, and oscillometry parameters of 20 patients aged 14–18 years who reached the appropriate tidal volume (VT) during MBNW.ResultsStatistical analysis showed a negative correlation between lung clearance index (LCI) and spimoetric parameters. Comparison of subgroups IA (n = 102) and IB (n = 48) indicated a statistically significant difference in LCI (p < 0.001) and FEV1z-score (p < 0.001), FEV1% pred (p < 0.001), MEF25z-score (p < 0.001), MEF50 z-score (p < 0.001), MEF75 z-score (p < 0.001), R5% pred (p < 0.05), and R20% pred (p < 0.01). LCI higher than 7.91 was found in 75.33% of the study group, in subgroup IB—91.67%, and IA−67.6%.ConclusionsLCI derived from MBNW may be a better tool than IOS for assessing pulmonary function in patients with CF, particularly those who cannot perform spirometry.

2022 ◽  
pp. 096703352110572
Nicholas T Anderson ◽  
Kerry B Walsh

Short wave near infrared (NIR) spectroscopy operated in a partial or full transmission geometry and a point spectroscopy mode has been increasingly adopted for evaluation of quality of intact fruit, both on-tree and on-packing lines. The evolution in hardware has been paralleled by an evolution in the modelling techniques employed. This review documents the range of spectral pre-treatments and modelling techniques employed for this application. Over the last three decades, there has been a shift from use of multiple linear regression to partial least squares regression. Attention to model robustness across seasons and instruments has driven a shift to machine learning methods such as artificial neural networks and deep learning in recent years, with this shift enabled by the availability of large and diverse training and test sets.

2022 ◽  
Vol 11 ◽  
Adrián Mosquera Orgueira ◽  
Miguel Cid López ◽  
Andrés Peleteiro Raíndo ◽  
Aitor Abuín Blanco ◽  
Jose Ángel Díaz Arias ◽  

Follicular Lymphoma (FL) has a 10-year mortality rate of 20%, and this is mostly related to lymphoma progression and transformation to higher grades. In the era of personalized medicine it has become increasingly important to provide patients with an optimal prediction about their expected outcomes. The objective of this work was to apply machine learning (ML) tools on gene expression data in order to create individualized predictions about survival in patients with FL. Using data from two different studies, we were able to create a model which achieved good prediction accuracies in both cohorts (c-indexes of 0.793 and 0.662 in the training and test sets). Integration of this model with m7-FLIPI and age rendered high prediction accuracies in the test set (cox c-index 0.79), and a simplified approach identified 4 groups with remarkably different outcomes in terms of survival. Importantly, one of the groups comprised 27.35% of patients and had a median survival of 4.64 years. In summary, we have created a gene expression-based individualized predictor of overall survival in FL that can improve the predictions of the m7-FLIPI score.

2022 ◽  
Vol 11 ◽  
Minghao Wu ◽  
Yanyan Zhang ◽  
Jianing Zhang ◽  
Yuwei Zhang ◽  
Yina Wang ◽  

ObjectiveBased on non-contrast-enhanced (NCE)/contrast-enhanced (CE) computed tomography (CT) images, we try to identify a combined-radiomics model and evaluate its predictive capacity regarding response to anti-PD1 immunotherapy of patients with non-small-cell lung cancer (NSCLC).Methods131 patients with NSCLC undergoing anti-PD1 immunotherapy were retrospectively enrolled from 7 institutions. Using largest lesion (LL) and target lesions (TL) approaches, we performed a radiomics analysis based on pretreatment NCE-CT (NCE-radiomics) and CE-CT images (CE-radiomics), respectively. Meanwhile, a combined-radiomics model based on NCE-CT and CE-CT images was constructed. Finally, we developed their corresponding nomograms incorporating clinical factors. ROC was used to evaluate models’ predictive performance in the training and testing set, and a DeLong test was employed to compare the differences between different models.ResultsFor TL approach, both NCE-radiomics and CE-radiomics performed poorly in predicting response to immunotherapy. For LL approach, NCE-radiomics nomograms and CE-radiomics nomograms incorporating with clinical factor of distant metastasis all showed satisfactory results, reflected by the AUCs in the training (AUC=0.84, 95% CI: 0.75-0.92; AUC=0.77, 95% CI: 0.67-0.87) and test sets (AUC=0.78, 95% CI: 0.64-0.92, AUC=0.73, 95% CI: 0.57-0.88), respectively. Compared with the NCE-radiomics nomograms, the combined-radiomics nomogram showed incremental predictive capacity in the training set (AUC=0.85, 95% CI: 0.77-0.92) and test set (AUC=0.81, 95% CI: 0.67-0.94), respectively, but no statistical difference (P=0.86, P=0.79).ConclusionCompared with radiomics based on single NCE or CE-CT images, the combined-radiomics model has potential advantages to identify patients with NSCLC most likely to benefit from immunotherapy, and may effectively improve more precise and individualized decision support.

2022 ◽  
pp. 1-17
Connor T. Jerzak ◽  
Gary King ◽  
Anton Strezhnev

Abstract Some scholars build models to classify documents into chosen categories. Others, especially social scientists who tend to focus on population characteristics, instead usually estimate the proportion of documents in each category—using either parametric “classify-and-count” methods or “direct” nonparametric estimation of proportions without individual classification. Unfortunately, classify-and-count methods can be highly model-dependent or generate more bias in the proportions even as the percent of documents correctly classified increases. Direct estimation avoids these problems, but can suffer when the meaning of language changes between training and test sets or is too similar across categories. We develop an improved direct estimation approach without these issues by including and optimizing continuous text features, along with a form of matching adapted from the causal inference literature. Our approach substantially improves performance in a diverse collection of 73 datasets. We also offer easy-to-use software that implements all ideas discussed herein.

2022 ◽  
Vol 12 ◽  
Yue Gong ◽  
Benzhi Dong ◽  
Zixiao Zhang ◽  
Yixiao Zhai ◽  
Bo Gao ◽  

Vesicular transport proteins are related to many human diseases, and they threaten human health when they undergo pathological changes. Protein function prediction has been one of the most in-depth topics in bioinformatics. In this work, we developed a useful tool to identify vesicular transport proteins. Our strategy is to extract transition probability composition, autocovariance transformation and other information from the position-specific scoring matrix as feature vectors. EditedNearesNeighbours (ENN) is used to address the imbalance of the data set, and the Max-Relevance-Max-Distance (MRMD) algorithm is adopted to reduce the dimension of the feature vector. We used 5-fold cross-validation and independent test sets to evaluate our model. On the test set, VTP-Identifier presented a higher performance compared with GRU. The accuracy, Matthew’s correlation coefficient (MCC) and area under the ROC curve (AUC) were 83.6%, 0.531 and 0.873, respectively.

2022 ◽  
Vol 12 (1) ◽  
pp. 0-0

Liver cancer is one the most common forms of cancer. As per statistics in 2018 published by World Health Organization, a quarter of all cancer cases are caused by infections, particularly prevalent in developing countries, including hepatitis B, which is linked to liver cancer. The mortality rate is higher in liver cancer as compared to other types of cancer. Quick and reliable diagnosis tools are of paramount importance for detecting and treating liver cancer in early stage, thus improving the likely course of a medical condition of patient. We have developed a cloud-based solution for liver tumour Segmentation, Classification and Detection in CT images based on GoogleNet architecture of Convolutional Neural Network. Experiment is carried out with training and test sets derived from TCIA repository. The results yield 96.7% accuracy for classification of tumour cells. GoogleNet architecture is used for implementation. The GoogleNet has 70,000 images in diagnosis of malignant tumor in liver cancer, providing a rich database for testing. Our algorithm has been deployed in Azure cloud.

10.6036/10117 ◽  
2022 ◽  
Vol 97 (1) ◽  
pp. 35-38

The popularity of the use of computational tools such as artificial intelligence has been increasing in recent years, and its importance in medicine is a fact. This field has benefited greatly thanks to the incorporation of more effective and faster methodologies in the medical diagnosis and registration processes. In the present work, the classification of images related to three diseases: Tuberculosis, Glaucoma and Parkinson's is carried out. We used deep learning and the RESNET50 convolutional neural network to extract classification characteristics, and then perform the classification based on standard methods, such as support vector machines, Naïve Bayes, and Centroid-based classifier, which are incorporated into two scenarios (cross validation; training and test sets). The classifier's performance is evaluated quantitatively using three evaluation metrics. The results obtained support the feasibility of the proposed methodology and its potential to improve medical diagnosis.

Rui Zhou ◽  
Yu-fang Liang ◽  
Hua-Li Cheng ◽  
Wei Wang ◽  
Da-wei Huang ◽  

Abstract Objectives Delta check (DC) is widely used for detecting sample mix-up. Owing to the inadequate error detection and high false-positive rate, the implementation of DC in real-world settings is labor-intensive and rarely capable of absolute detection of sample mix-ups. The aim of the study was to develop a highly accurate DC method based on designed deep learning to detect sample mix-up. Methods A total of 22 routine hematology test items were adopted for the study. The hematology test results, collected from two hospital laboratories, were independently divided into training, validation, and test sets. By selecting six mainstream algorithms, the Deep Belief Network (DBN) was able to learn error-free and artificially (intentionally) mixed sample results. The model’s analytical performance was evaluated using training and test sets. The model’s clinical validity was evaluated by comparing it with three well-recognized statistical methods. Results When the accuracy of our model in the training set reached 0.931 at the 22nd epoch, the corresponding accuracy in the validation set was equal to 0.922. The loss values for the training and validation sets showed a similar (change) trend over time. The accuracy in the test set was 0.931 and the area under the receiver operating characteristic curve was 0.977. DBN demonstrated better performance than the three comparator statistical methods. The accuracy of DBN and revised weighted delta check (RwCDI) was 0.931 and 0.909, respectively. DBN performed significantly better than RCV and EDC. Of all test items, the absolute difference of DC yielded higher accuracy than the relative difference for all methods. Conclusions The findings indicate that input of a group of hematology test items provides more comprehensive information for the accurate detection of sample mix-up by machine learning (ML) when compared with a single test item input method. The DC method based on DBN demonstrated highly effective sample mix-up identification performance in real-world clinical settings.

Sign in / Sign up

Export Citation Format

Share Document