Multifactor Prediction of Embryo Transfer Outcomes Based on a Machine Learning Algorithm

In vitro fertilization-embryo transfer (IVF-ET) technology make it possible for infertile couples to conceive a baby successfully. Nevertheless, IVF-ET does not guarantee success. Frozen embryo transfer (FET) is an important supplement to IVF-ET. Many factors are correlated with the outcome of FET which is unpredictable. Machine learning is a field of study that predict various outcomes by defining data attributes and using relevant data and calculation algorithms. Machine learning algorithm has been widely used in clinical research. The present study focuses on making predictions of early pregnancy outcomes in FET through clinical characters, including age, body mass index (BMI), endometrial thickness (EMT) on the day of progesterone treatment, good-quality embryo rate (GQR), and type of infertility (primary or secondary), serum estradiol level (E2) on the day of embryo transfer, and serum progesterone level (P) on the day of embryo transfer. We applied four representative machine learning algorithms, including logistic regression (LR), conditional inference tree, random forest (RF) and support vector machine (SVM) to build prediction models and identify the predictive factors. We found no significant difference among the models in the sensitivity, specificity, positive predictive rate, negative predictive rate or accuracy in predicting the pregnancy outcome of FET. For example, the positive/negative predictive rate of the SVM (gamma = 1, cost = 100, 10-fold cross validation) is 0.56 and 0.55. This approach could provide a reference for couples considering FET. The prediction accuracy of the present study is limited, which suggests that there may be some other more effective predictors to be developed in future work.

Predicting Aphasia from Strokes

Strokes can occur when someone’s blood vessels get blocked and the nutrients and oxygen being transported will not reach the brain. When a stroke happens, the brain cells don’t get the nutrients they need and start to die [3]. This could cause different side effects after stroke. In this study, we try to predict the possibility of one type of after-stroke side effect, aphasia, using Machine Learning (ML) techniques. Using the data of a study about brain lesion damage after a stroke and what effects the patients were experiencing afterward, we trained a model to predict whether a person may have aphasia based on where their lesion was, how big the lesion was, how long ago their stroke was, and some other factors. We evaluated several classification methods and found that using linear discriminant analysis was the most accurately predicting when we used age, sex, lesion location, lesion volume, and many more. By linear discriminant analysis, we were able to have a 91% overall predictive rate of patients having aphasia or not after experiencing a stroke.

An Effective Approach Based on Deep Residual Google Net Convolutional Neural Network Classifier for the Detection of Glaucoma

Glaucoma is a disease caused by fluid pressure build-up in the inner eye. Early detection of glaucoma is critical as it is expected that 111.8 million people worldwide shall suffer from glaucoma in 2040. In the diagnosis of glaucoma, the use of machine learning method is hoped to be highly promising. This paper provides an important method to master learning to diagnose glaucoma. Initially, human retinal fundus images are preprocessed by means of histogram equalization in order to enhance them. The segmentation is performed by semantic segmentation method, mainly the features are extracted using density with correlation based feature extraction approach. PCA (principal component analysis) methodology is used to choose the most optimal features. Ultimately, through the usage of the Deep residual Google Net CNN Classification method, the retinal image is classified/predicted as regular and abnormal. The Deep residual Google Net CNN classifier is designed to distinguish view patterns with minimal pre-processing from pixel pictures. ORIGA and STARE datasets are used in this work. The findings are then analyzed and contrasted to illustrate the efficacy of the new technique with alternate current techniques. Test accuracy of 99%, Specificity of 98.9% and 100% Sensitivity were achieved. The quantitative results are analyzed for specifications like sensitivity, specificity, accuracy, positive predictive rate, false predictive rate and assured to provide excellent outcomes when compared with traditional methods.

Taiwanese Clinical Experience with Noninvasive Prenatal Testing for DiGeorge Syndrome

<b><i>Objective:</i></b> DiGeorge syndrome (DGS) is associated with microdeletions of chromosome 22q11. It is the second most common cause of congenital heart disease and is an important consideration whenever a conotruncal cardiac anomaly is identified. The availability of noninvasive prenatal testing (NIPT) is altering the practice of prenatal genetics and maternal-fetal medicine, resulting in a decline in invasive testing. Antenatal ultrasound and other biomarkers have their own limitation. NIPT was proposed to screen DGS with cell-free DNA in Taiwan. Here, we present our experience of prenatal diagnosis of DGS in our center. <b><i>Methods:</i></b> This was a retrospective study between November 1, 2019, and August 31, 2020, in Taiwan. Data were collected from 7,826 pregnant women self-referred for DGS screening with massive parallel shotgun sequencing-based NIPT. High-risk cases subsequently received amniocentesis for array comparative genomic hybridization (aCGH) to confirm the diagnosis. Characteristics of pregnancies were documented when participants received the test. Report of NIPT was completed 2 weeks after the test. Follow-up on high-risk cases was completed by telephone interview on January 30, 2021. <b><i>Results:</i></b> Thirteen cases showed high risk by NIPT, and 7 cases were confirmed by aCGH. The sensitivity and specificity were 100% (95% confidence interval [CI] 64.57–100.00%) and 99.92% (95% CI 99.83–99.96%). The prevalence of DGS was 1 in 1,118 pregnancies. The positive predictive rate was 53.85% (95% CI 29.14–76.79%). One true positive (TP) showed US anomaly, and 5 TPs selected termination. <b><i>Discussion/Conclusion:</i></b> NIPT demonstrated good performance in DGS screening. Detection of 22q11.2 deletion could be combined with routine screening to facilitate proper intervention.

Is PSA density of the peripheral zone as a useful predictor for prostate cancer in patients with gray zone PSA levels?

Background Serum prostate-specific antigen (PSA) is widely used in screening tests for prostate cancer. As the low specificity of PSA results in unnecessary and invasive prostate biopsies, we evaluated the clinical significance of various PSAs and PSA density (PSAD) related to peripheral zones in patients with gray zone PSA level (4–10 ng/mL). Methods A total of 1300 patients underwent transrectal ultrasonography-guided prostate biopsy from 2014 to 2019. Among them, 545 patients in the gray zone were divided into the prostate cancer diagnosis group and the non-prostate cancer diagnosis group, and PSA, relative extra transitional zone PSA (RETzPSA), estimated post holmium laser enucleation of the prostate PSA (EPHPSA), PSAD, peripheral zone PSA density (PZPSAD) and extra-transitional zone density (ETzD) were compared and analyzed using receiver-operating characteristics (ROC) analysis after 1:1 matching using propensity score. Results Area under the ROC curve values of PSA, EPHPSA, RETzPSA, PSA density, ETzD, and PZPSAD were 0.553 (95% CI: 0.495–0.610), 0.611 (95% CI: 0.554–0.666), 0.673 (95% CI: 0.617–0.725), 0.745 (95% CI: 0.693–0.793), 0.731 (95% CI: 0.677–0.780) and 0.677 (95% CI: 0.611–0.719), respectively. PSAD had 67.11% sensitivity, 71.71% specificity, and 70.34% positive predictive rate at 0.18 ng/mL/cc. ETzD had 69.08% sensitivity, 64.47% specificity, and 66.04% positive predictive rate at 0.04 ng/mL/cc. When the cut-off value of PSAD was increased to 0.18 ng/mL/cc, the best results were obtained with an odds ratio of 5.171 (95% CI: 3.171–8.432), followed by ETzD with 4.054 (95% CI: 2.513–6.540). Conclusions These results suggested that volume-adjusted parameters (ETzD and PSAD) might be more sensitive and accurate than various PSA in gray zone patients who required prostate biopsy to reduce unnecessary biopsy.

DEM Resolutions for Landslide Susceptibility Modeling

In landslide susceptibility mapping, the digital elevation model (DEM) is one of the most essential data sets, which is frequently used. Therefore, evaluate the effects of the spatial resolution of DEM on the landslide susceptibility model is very important. Hence, this paper is analyzed only the effects of the spatial resolution of DEM, Advanced Spaceborne Thermal Emission, and Reflection (ASTER) was used for DEM data source. The ASTER DEM was resampled to 45, 60, 75, and 90 m spatial resolutions. A set of geodatabases were built using Geographic Information System (GIS), which contains landslide governing factors and landslide inventory. Frequency ratio (FR) and certainty factor (CF) statistical methods were employed to generate a landslide susceptibility map. Landslide density and area under the curve (AUC) were applied to evaluate the model's performance for each DEM resolution. The results of the predictive rate curve value of AUC showed a coarser DEM resolution (90 m) produced the best performance and prediction accuracy. This indicated that a coarser DEM resolution produced higher predictive accuracy than fine resolution. Concerning the statistical models, the frequency ratio model produced very good accuracy at the coarser DEM resolutions (75 and 90 m). The predictive rate curve value of AUC ranges from 86-92% for the FR model and 81-89% for the CF model which indicating very good accuracy of the models to predict future landslide incidence in the study area. Therefore, it is possible to endorse statistical methods (frequency ratio, and certainty factor) respect with to DEM resolution, which is satisfactory to landslide susceptibility mapping.

DEM Resolutions for Landslide Susceptibility Modeling

In landslide susceptibility mapping, the digital elevation model (DEM) is one of the most essential data sets, which is frequently used. Therefore, evaluate the effects of the spatial resolution of DEM on the landslide susceptibility model is very important. Hence, this paper is analyzed only the effects of the spatial resolution of DEM, Advanced Spaceborne Thermal Emission, and Reflection (ASTER) was used for DEM data source. The ASTER DEM was resampled to 45, 60, 75, and 90 m spatial resolutions. A set of geodatabases were built using Geographic Information System (GIS), which contains landslide governing factors and landslide inventory. Frequency ratio (FR) and certainty factor (CF) statistical methods were employed to generate a landslide susceptibility map. Landslide density and area under the curve (AUC) were applied to evaluate the model's performance for each DEM resolution. The results of the predictive rate curve value of AUC showed a coarser DEM resolution (90 m) produced the best performance and prediction accuracy. This indicated that a coarser DEM resolution produced higher predictive accuracy than fine resolution. Concerning the statistical models, the frequency ratio model produced very good accuracy at the coarser DEM resolutions (75 and 90 m). The predictive rate curve value of AUC ranges from 86–92% for the FR model and 81–89% for the CF model which indicating very good accuracy of the models to predict future landslide incidence in the study area. Therefore, it is possible to endorse statistical methods (frequency ratio, and certainty factor) respect with to DEM resolution, is satisfactory to landslide susceptibility mapping.

Automated detection of mouse scratching behaviour using convolutional recurrent neural network

Scratching is one of the most important behaviours in experimental animals because it can reflect itching and/or psychological stress. Here, we aimed to establish a novel method to detect scratching using deep neural network. Scratching was elicited by injecting a chemical pruritogen lysophosphatidic acid to the back of a mouse, and behaviour was recorded using a standard handy camera. Images showing differences between two consecutive frames in each video were generated, and each frame was manually labelled as showing scratching behaviour or not. Next, a convolutional recurrent neural network (CRNN), composed of sequential convolution, recurrent, and fully connected blocks, was constructed. The CRNN was trained using the manually labelled images and then evaluated for accuracy using a first-look dataset. Sensitivity and positive predictive rates reached 81.6% and 87.9%, respectively. The predicted number and durations of scratching events correlated with those of the human observation. The trained CRNN could also successfully detect scratching in the hapten-induced atopic dermatitis mouse model (sensitivity, 94.8%; positive predictive rate, 82.1%). In conclusion, we established a novel scratching detection method using CRNN and showed that it can be used to study disease models.

Is PSA Density of the Peripheral Zone as a Useful Predictor for Prostate Cancer in Patients With Gray Zone PSA Levels?

BackgroundSerum prostate-specific antigen (PSA) is widely used in screening tests for prostate cancer. As the low specificity of PSA results in unnecessary and invasive prostate biopsies, we evaluated the clinical significance of various PSAs and PSA density (PSAD) related to peripheral zones in patients with gray zone PSA level (4-10 ng/mL).MethodsA total of 1,300 patients underwent transrectal ultrasonography-guided prostate biopsy from 2014 to 2019. Among them, 545 patients in the gray zone were divided into the prostate cancer diagnosis group and the non-prostate cancer diagnosis group, and PSA, relative extra transitional zone PSA (RETzPSA), estimated post holmium laser enucleation of the prostate PSA (EPHPSA), PSAD, peripheral zone PSA density (PZPSAD) and extra-transitional zone density (ETzD) were compared and analyzed using receiver-operating characteristics (ROC) analysis after 1:1 matching using propensity score.ResultsArea under the ROC curve values of PSA, EPHPSA, RETzPSA, PSA density, ETzD, and PZPSAD were 0.553 (95% CI: 0.495-0.610), 0.611 (95% CI: 0.554-0.666), 0.673 (95% CI: 0.617-0.725), 0.745 (95% CI: 0.693-0.793), 0.731 (95% CI: 0.677-0.780) and 0.677 (95% CI: 0.611-0.719), respectively. PSAD had 67.11% sensitivity, 71.71% specificity, and 70.34% positive predictive rate at 0.18 ng/mL/cc. ETzD had 69.08% sensitivity, 64.47% specificity, and 66.04% positive predictive rate at 0.04 ng/mL/cc. When the cut-off value of PSAD was increased to 0.18 ng/mL/cc, the best results were obtained with an odds ratio of 5.171 (95% CI: 3.171-8.432), followed by ETzD with 4.054 (95% CI: 2.513-6.540).ConclusionsThese results suggested that volume-adjusted parameters (ETzD and PSAD) might be more sensitive and accurate than various PSA in gray zone patients who required prostate biopsy to reduce unnecessary biopsy.