Deep-learning based discrimination of pathologic complete response using MRI in HER2-positive and triple-negative breast cancer.

Distinguishing between pathologic complete response and residual cancer after neoadjuvant chemotherapy (NAC) is crucial for treatment decisions, but the current imaging methods face challenges. To address this, we developed deep-learning models using post-NAC dynamic contrast-enhanced MRI and clinical data. A total of 852 women with human epidermal growth factor receptor 2 (HER2)-positive or triple-negative breast cancer were randomly divided into a training set (n = 724) and a validation set (n = 128). A 3D convolutional neural network model was trained on the training set and validated independently. The main models were developed using cropped MRI images, but models using uncropped whole images were also explored. The delayed-phase model demonstrated superior performance compared to the early-phase model (area under the receiver operating characteristic curve [AUC] = 0.74 vs. 0.69, P = 0.013) and the combined model integrating multiple dynamic phases and clinical data (AUC = 0.74 vs. 0.70, P = 0.022). Deep-learning models using uncropped whole images exhibited inferior performance, with AUCs ranging from 0.45 to 0.54. Further refinement and external validation are necessary for enhanced accuracy.

Identification of Preeclamptic Placenta in Whole Slide Images Using Artificial Intelligence Placenta Analysis.

BACKGROUND: Preeclampsia (PE) is a hypertensive pregnancy disorder linked to placental dysfunction, often involving pathological lesions like acute atherosis, decidual vasculopathy, accelerated villous maturation, and fibrinoid deposition. However, there is no gold standard for the pathological diagnosis of PE and this limits the ability of clinicians to distinguish between PE and non-PE pregnancies. Recent advances in computational pathology have provided the opportunity to automate pathological analysis for diagnosis, classification, prediction, and prediction of disease progression. In this study, we assessed whether computational pathology could be used to identify PE placentas. METHODS: A total of 168 placental whole-slide images (WSIs) of patients from Seoul National University Hospital (comprising 84 PE cases and 84 normal controls) were used for model development and internal validation. For external validation of the model, 76 placental slides (including 38 PE cases and 38 normal controls) were obtained from the Boramae Medical Center (BMC). To establish standard criteria for diagnosing PE and distinguishing it from controls using placental WSIs, patch characteristics and quantification of terminal and intermediate villi were employed. In unsupervised learning, K-means clustering was conducted as a feature obtained through an Auto Encoder to extract the ratio of each cluster for each WSI. For supervised learning, quantitative assessments of the villi were obtained using a U-Net-based segmentation algorithm. The prediction model was developed using an ensemble method and was compared with a clinical feature model developed by using placental size features. RESULTS: Using ensemble modeling, we developed a model to identify PE placentas. The model showed good performance (area under the precision-recall curve [AUPRC], 0.771; 95% confidence interval [CI], 0.752-0.790), with 77.3% of sensitivity and 71.1% of specificity, whereas the clinical feature model showed an AUPRC 0.713 (95% CI, 0.694-0.732) with 55.6% sensitivity and 86.8% specificity. External validation of the predictive model employing the BMC-derived set of placental slides also showed good discrimination (AUPRC, 0.725; 95% CI, 0.720-0.730). CONCLUSION: The proposed computational pathology model demonstrated a strong ability to identify preeclamptic placentas. Computational pathology has the potential to improve the identification of PE placentas.

Conventional machine learning-based prediction models did not outperform the International IgA Nephropathy Prediction Tool.

Background: Immunoglobulin A nephropathy (IgAN) is a major cause of end-stage kidney disease (ESKD). The International IgA Nephropathy Prediction Tool (IIgAN-PT) predicts IgAN prognosis, but improvement in the prediction performance using machine learning (ML)-based methods is needed. Methods: We analyzed 4,425 biopsy-confirmed patients with IgAN and ≥6 months of follow-up from nine tertiary university hospitals in Korea. The study population was divided into development and validation cohorts. Using the collected 87 clinicodemographic and pathological variables, ML-based prediction models for ESKD or estimated glomerular filtration rate were constructed: 1) the conventional CatBoost model, 2) the optimized CatBoost model with Cox proportional hazards, 3) the deep Cox proportional hazards model, and 4) the deep Cox mixture model. The area under the curve (AUC) and calibration plots were used to investigate the discriminative and calibration performance of the models, which were then compared with those of the IIgAN-PT full model. Results: The full model showed excellent performance (AUC [95% confidence interval] for 5-year outcome, 0.896 [0.8530.940]), with acceptable calibration results. The ML-based models showed good performance in predicting adverse kidney outcomes and revealed acceptable discrimination performance in the external validation (AUC [95% confidence interval] for the 5-year outcome: 1) 0.829 [0.791-0.866]; 2) 0.847 [0.804-0.890]; 3) 0.823 [0.784-0.862]; and 4) 0.832 [0.794-0.870]), although they underestimated the external validation cohort risks. With the validation data, the overall performance of the IIgAN-PT was non-inferior to that of the ML-based model. Conclusions: Our ML-based models showed good performance in predicting adverse kidney outcomes in patients with IgAN but they did not outperform the IIgAN-PT.

Identification of diagnostic challenges in RP1 Alu insertion and strategies for overcoming them.

Recently, a founder Alu insertion in exon 4 of RP1 was detected in Japanese and Korean patients with inherited retinal diseases (IRDs). However, carrier frequency and diagnostic challenges for detecting AluY insertion are not established. We aim to investigate the frequency of AluY in individuals with or without IRDs and to overcome common diagnostic pitfalls associated with AluY insertion. A total of 1,072 subjects comprising 411 patients with IRD (IRD group) and 661 patients with other suspected Mendelian genetic disease (non-IRD group) was screened for AluY insertion. Targeted panel sequencing and whole-genome sequencing were used for detection of AluY insertion, and an optimized allele-specific PCR (AS-PCR) was used for validation. The AluY insertion was detected in 1.5% in IRD group (6/411). The AluY insertion was not observed in non-IRD group (0/661). All patients with AluY were confirmed to have RP1 pathogenic variants on the paired allele. We identified AluY allele dropout leading to false homozygosity for c.4196del pathogenic variant in Sanger sequencing. The allelic relationship between variants of RP1 was accurately determined by AluY AS-PCR. Delineating diagnostic challenges of AluY insertion and strategies to avoid potential pitfalls could aid clinicians in an accurate molecular diagnosis for patients with IRD.

Increased coherence predicts medical refractoriness in patients with temporal lobe epilepsy on monotherapy.

Among patients with epilepsy, 30-40% experience recurrent seizures even after adequate antiseizure medications therapies, making them refractory. The early identification of refractory epilepsy is important to provide timely surgical treatment for these patients. In this study, we analyze interictal electroencephalography (EEG) data to predict drug refractoriness in patients with temporal lobe epilepsy (TLE) who were treated with monotherapy at the time of the first EEG acquisition. Various EEG features were extracted, including statistical measurements and interchannel coherence. Feature selection was performed to identify the optimal features, and classification was conducted using different classifiers. Functional connectivity and graph theory measurements were calculated to identify characteristics of refractory TLE. Among the 48 participants, 34 (70.8%) were responsive, while 14 (29.2%) were refractory over a mean follow-up duration of 38.5 months. Coherence feature within the gamma frequency band exhibited the most favorable performance. The light gradient boosting model, employing the mutual information filter-based feature selection method, demonstrated the highest performance (AUROC = 0.821). Compared to the responsive group, interchannel coherence displayed higher values in the refractory group. Interestingly, graph theory measurements using EEG coherence exhibited higher values in the refractory group than in the responsive group. Our study has demonstrated a promising method for the early identification of refractory TLE utilizing machine learning algorithms.

Electrospun Silicon Dioxide/poly(vinylidene fluoride) Nanofibrous Membrane Comprising a Skin Multicore-Shell Nanostructure as a New High-Heat-Resistant Separator for Lithium-Ion Polymer Batteries.

Porous silicon dioxide (SiO2)/poly(vinylidene fluoride) (PVdF), SiO2/PVdF, and fibrous composite membranes were prepared by electrospinning a blend solution of a SiO2 sol-gel/PVdF. The nanofibers of the SiO2/PVdF (3/7 wt. ratio) blend comprised skin and nanofibrillar structures which were obtained from the SiO2 component. The thickness of the SiO2 skin layer comprising a thin skin layer could be readily tuned depending on the weight proportions of SiO2 and PVdF. The composite membrane exhibited a low thermal shrinkage of ~3% for 2 h at 200 °C. In the prototype cell comprising the composite membrane, the alternating current impedance increased rapidly at ~225 °C, and the open-circuit voltage steeply decreased at ~170 °C, almost becoming 0 V at ~180 °C. After being exposed at temperatures of >270 °C, its three-dimensional network structure was maintained without the closure of the pore structure by a melt-down of the membrane.

Effects of Korean Versus Western Diets on Reproductive Function in Young Korean Men: A 12-Week Randomized Parallel Clinical Trial.

Background: Studies report that diet may have contributed to a 50-60% decrease in human sperm quality over the past few decades. Unhealthy lifestyles affect the structure of spermatozoa, affecting the male reproductive potential. This study aimed to compare the effects of Korean and Western diets on reproductive function in young male Koreans. Methods: Study participants were provided either the Korean Diet (KD group) or the Western Diet (WD group) for 12 weeks. Semen quality parameters such as volume, motility, cell count, and sex hormone levels were evaluated. Sexual function was assessed using the International Index of Erectile Function and the Male Sexual Health Questionnaire. Efficacy and safety evaluations were conducted at baseline, 8 weeks, and 12 weeks. Results: The KD group demonstrated a significantly increased sperm motility after 8 weeks relative to baseline but decreased after 12 weeks. In contrast, sperm motility in the WD group significantly decreased after 8 weeks compared with baseline and remained constant after 12 weeks. Statistically, a near-significant difference was observed between groups (p = 0.057). Similarly, free testosterone levels in the KD group increased after 12 weeks compared with baseline, whereas that in the WD group decreased. The free testosterone levels in the KD group were significantly higher than those in the WD group (p = 0.020). There were no statistically significant differences in other sex hormone and sexual function questionnaires between the groups. None of the participants reported any severe side effects, and no significant alterations in clinical diagnostic test values were detected. Conclusion: The results of the study strongly reveal that KD positively affects sperm motility and male hormone levels in young men, indicating potential benefits for reproductive function.

Clinical feature, GALC variant spectrum, and genotype-phenotype correlation in Korean Krabbe disease patients: Multicenter experience over 13 years.

Krabbe disease (KD) is an autosomal recessive neurodegenerative disorder caused by deficiency of the galactocerebrosidase (GALC) due to variants in the GALC gene. Here, we provide the first and the largest comprehensive analysis of clinical and genetic characteristics, and genotype-phenotype correlations of KD in Korean in comparison with other ethnic groups. From June 2010 to June 2023, 10 patients were diagnosed with KD through sequencing of GALC. Clinical features, and results of GALC sequencing, biochemical test, neuroimaging, and neurophysiologic test were obtained from medical records. An additional nine previously reported Korean KD patients were included for review. In Korean KD patients, the median age of onset was 2 years (3 months-34 years) and the most common phenotype was adult-onset (33%, 6/18) KD, followed by infantile KD (28%, 5/18). The most frequent variants were c.683_694delinsCTC (23%) and c.1901T>C (23%), while the 30-kb deletion was absent. Having two heterozygous pathogenic missense variants was associated with later-onset phenotype. Clinical features were similar to those of other ethnic groups. In Korean KD patients, the most common phenotype was the adult-onset type and the GALC variant spectrum was different from that of the Caucasian population. This study would further our understanding of KD.

Deep learning-based fully automated grading system for dry eye disease severity.

There is an increasing need for an objective grading system to evaluate the severity of dry eye disease (DED). In this study, a fully automated deep learning-based system for the assessment of DED severity was developed. Corneal fluorescein staining (CFS) images of DED patients from one hospital for system development (n = 1400) and from another hospital for external validation (n = 94) were collected. Three experts graded the CFS images using NEI scale, and the median value was used as ground truth. The system was developed in three steps: (1) corneal segmentation, (2) CFS candidate region classification, and (3) estimation of NEI grades by CFS density map generation. Also, two images taken on different days in 50 eyes (100 images) were compared to evaluate the probability of improvement or deterioration. The Dice coefficient of the segmentation model was 0.962. The correlation between the system and the ground truth data was 0.868 (p<0.001) and 0.863 (p<0.001) for the internal and external validation datasets, respectively. The agreement rate for improvement or deterioration was 88% (44/50). The fully automated deep learning-based grading system for DED severity can evaluate the CFS score with high accuracy and thus may have potential for clinical application.

Enhancing the Reliability of PMP22 Copy Number Variation Detection with an Inherited Peripheral Neuropathy Panel.

The utility of the next-generation sequencing (NGS) panel could be increased in hereditary peripheral neuropathies, given that the duplication of PMP22 is a major abnormality. In the present study, the analytical performance of an algorithm for detecting PMP22 copy number variation (CNV) from the NGS panel data was evaluated. The NGS panel covers 141 genes, including PMP22 and five genes within 1.5-megabase duplicated region at 17p11.2. CNV calling was performed using a laboratory-developed algorithm. Among the 92 cases subjected to targeted NGS panel from March 2018 to January 2021, 26 were suggestive of PMP22 CNV. Multiplex ligation-dependent probe amplification analysis was performed in 58 cases, and the results were 100% concordant with the NGS data (23 duplications, 2 deletions, and 33 negatives). Analytical performance of the pipeline was further validated by another blind data set, including 14 positive and 20 negative samples. Reliable detection of PMP22 CNV was possible by analyzing not only PMP22 but also the adjacent genes within the 1.5-megabase region of 17p11.2. On the basis of the high accuracy of CNV calling for PMP22, the testing strategy for diagnosis of peripheral polyneuropathies could be simplified by reducing the need for multiplex ligation-dependent probe amplification.

A CT-based Deep Learning Model for Predicting Subsequent Fracture Risk in Patients with Hip Fracture.

Background Patients have the highest risk of subsequent fractures in the first few years after an initial fracture, yet models to predict short-term subsequent risk have not been developed. Purpose To develop and validate a deep learning prediction model for subsequent fracture risk using digitally reconstructed radiographs from hip CT in patients with recent hip fractures. Materials and Methods This retrospective study included adult patients who underwent three-dimensional hip CT due to a fracture from January 2004 to December 2020. Two-dimensional frontal, lateral, and axial digitally reconstructed radiographs were generated and assembled to construct an ensemble model. DenseNet modules were used to calculate risk probability based on extracted image features and fracture-free probability plots were output. Model performance was assessed using the C index and area under the receiver operating characteristic curve (AUC) and compared with other models using the paired t test. Results The training and validation set included 1012 patients (mean age, 74.5 years ± 13.3 [SD]; 706 female, 113 subsequent fracture) and the test set included 468 patients (mean age, 75.9 years ± 14.0; 335 female, 22 subsequent fractures). In the test set, the ensemble model had a higher C index (0.73) for predicting subsequent fractures than that of other image-based models (C index range, 0.59-0.70 for five of six models; P value range, < .001 to < .05). The ensemble model achieved AUCs of 0.74, 0.74, and 0.73 at the 2-, 3-, and 5-year follow-ups, respectively; higher than that of most other image-based models at 2 years (AUC range, 0.57-0.71 for five of six models; P value range, < .001 to < .05) and 3 years (AUC range, 0.55-0.72 for four of six models; P value range, < .001 to < .05). Moreover, the AUCs achieved by the ensemble model were higher than that of a clinical model that included known risk factors (2-, 3-, and 5-year AUCs of 0.58, 0.64, and 0.70, respectively; P < .001 for all). Conclusion In patients with recent hip fractures, the ensemble deep learning model using digital reconstructed radiographs from hip CT showed good performance for predicting subsequent fractures in the short term. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Li and Jaremko in this issue.

Clinical Practice Guideline for Blood-based Circulating Tumor DNA Assays.

Circulating tumor DNA (ctDNA) has emerged as a promising tool for various clinical applications, including early diagnosis, therapeutic target identification, treatment response monitoring, prognosis evaluation, and minimal residual disease detection. Consequently, ctDNA assays have been incorporated into clinical practice. In this review, we offer an in-depth exploration of the clinical implementation of ctDNA assays. Notably, we examined existing evidence related to pre-analytical procedures, analytical components in current technologies, and result interpretation and reporting processes. The primary objective of this guidelines is to provide recommendations for the clinical utilization of ctDNA assays.

Overcoming challenges associated with identifying FBN1 deep intronic variants through whole-genome sequencing.

BACKGROUND: Marfan syndrome (MFS), caused by pathogenic variants of FBN1 (fibrillin-1), is a systemic connective tissue disorder with variable phenotypes and treatment responsiveness depending on the variant. However, a significant number of individuals with MFS remain genetically unexplained. In this study, we report novel pathogenic intronic variants in FBN1 in two unrelated families with MFS. METHODS: We evaluated subjects with suspected MFS from two unrelated families using Sanger sequencing or multiplex ligation-dependent probe amplification of FBN1 and/or panel-based next-generation sequencing. As no pathogenic variants were identified, whole-genome sequencing was performed. Identified variants were analyzed by reverse transcription-PCR and targeted sequencing of FBN1 mRNA harvested from peripheral blood or skin fibroblasts obtained from affected probands. RESULTS: We found causative deep intronic variants, c.6163+1484A>T and c.5788+36C>A, in FBN1. The splicing analysis revealed an insertion of in-frame or out-of-frame intronic sequences of the FBN1 transcript predicted to alter function of calcium-binding epidermal growth factor protein domain. Family members carrying c.6163+1484A>T had high systemic scores including prominent skeletal features and aortic dissection with lesser aortic dilatation. Family members carrying c.5788+36C>A had more severe aortic root dilatation without aortic dissection. Both families had ectopia lentis. CONCLUSION: Variable penetrance of the phenotype and negative genetic testing in MFS families should raise the possibility of deep intronic FBN1 variants and the need for additional molecular studies. This study expands the mutation spectrum of FBN1 and points out the importance of intronic sequence analysis and the need for integrative functional studies in MFS diagnosis.

The epidemiology of male lower urinary tract symptoms associated with benign prostatic hyperplasia: Results of 20 years of Korean community care and surveys.

PURPOSE: To investigate the prevalence of lower urinary tract symptoms/benign prostatic hyperplasia in a Korean population. MATERIALS AND METHODS: The Korean Prostate & Voiding Health Association provided free prostate-related community health care and conducted surveys in all regions of Korea from 2001 to 2022 with the cooperation of local government public health centers. A total of 72,068 males older than 50 were surveyed and analyzed. History taking, International Prostate Symptom Score (IPSS), transrectal ultrasonography, prostate-specific antigen (PSA) testing, uroflowmetry, and urine volume testing were performed. RESULTS: The mean prostate volumes in males in their 50s, 60s, 70s, and 80s or above were 24.7 g, 27.7 g, 31 g, and 33.7 g, respectively. The proportion of males with high PSA greater than 3 ng/mL was 3.8% among males in their 50s, 7.7% among males in their 60s, 13.1% among males in their 70s, and 17.9% among males 80 years of age or older. The mean IPSS total scores in males in their 50s, 60s, 70s, and 80s or above were 10.7, 12.7, 14.5, and 16, respectively. Severe symptoms were reported by 27.3% of males, whereas 51.7% reported moderate symptoms. The mean Qmax in males in their 50s, 60s, 70s, and 80s or above were 20 mL/s, 17.4 mL/s, 15.4 mL/s, and 13.8 mL/s, respectively. CONCLUSIONS: In this population-based study, mean prostate volume, IPSS, PSA, and Qmax were 30.6±15.1 g, 14.8±8.2, 1.9±4.7 ng/mL, and 15.6±6.5 mL/s, respectively. Aging was significantly associated with increased prostate volume, PSA levels, and IPSS scores, and with decreased Qmax and urine volume.

CRYAB stop-loss variant causes rare syndromic dilated cardiomyopathy with congenital cataract: expanding the phenotypic and mutational spectrum of alpha-B crystallinopathy.

Missense mutations in the alpha-B crystallin gene (CRYAB) have been reported in desmin-related myopathies with or without cardiomyopathy and have also been reported in families with only a cataract phenotype. Dilated cardiomyopathy (DCM) is a disorder with a highly heterogeneous genetic etiology involving more than 60 causative genes, hindering genetic diagnosis. In this study, we performed whole genome sequencing on 159 unrelated patients with DCM and identified an unusual stop-loss pathogenic variant in NM_001289808.2:c.527A>G of CRYAB in one patient. The mutant alpha-B crystallin protein is predicted to have an extended strand with addition of 19 amino acid residues, p.(Ter176TrpextTer19), which may contribute to aggregation and increased hydrophobicity of alpha-B crystallin. The proband, diagnosed with DCM at age 32, had a history of bilateral congenital cataracts but had no evidence of myopathy or associated symptoms. He also has a 10-year-old child diagnosed with bilateral congenital cataracts with the same CRYAB variant. This study expands the mutational spectrum of CRYAB and deepens our understanding of the complex phenotypes of alpha-B crystallinopathies.

Multi-pose-based convolutional neural network model for diagnosis of patients with central lumbar spinal stenosis.

Although the role of plain radiographs in diagnosing lumbar spinal stenosis (LSS) has declined in importance since the advent of magnetic resonance imaging (MRI), diagnostic ability of plain radiographs has improved dramatically when combined with deep learning. Previously, we developed a convolutional neural network (CNN) model using a radiograph for diagnosing LSS. In this study, we aimed to improve and generalize the performance of CNN models and overcome the limitation of the single-pose-based CNN (SP-CNN) model using multi-pose radiographs. Individuals with severe or no LSS, confirmed using MRI, were enrolled. Lateral radiographs of patients in three postures were collected. We developed a multi-pose-based CNN (MP-CNN) model using the encoders of the three SP-CNN model (extension, flexion, and neutral postures). We compared the validation results of the MP-CNN model using four algorithms pretrained with ImageNet. The MP-CNN model underwent additional internal and external validations to measure generalization performance. The ResNet50-based MP-CNN model achieved the largest area under the receiver operating characteristic curve (AUROC) of 91.4% (95% confidence interval [CI] 90.9-91.8%) for internal validation. The AUROC of the MP-CNN model were 91.3% (95% CI 90.7-91.9%) and 79.5% (95% CI 78.2-80.8%) for the extra-internal and external validation, respectively. The MP-CNN based heatmap offered a logical decision-making direction through optimized visualization. This model holds potential as a screening tool for LSS diagnosis, offering an explainable rationale for its prediction.

Differential Diagnosis of Pleural Effusion Using Machine Learning.

Rationale: Differential diagnosis of pleural effusion is challenging in clinical practice. Objectives: We aimed to develop a machine learning model to classify the five common causes of pleural effusions. Methods: This retrospective study collected 49 features from clinical information, blood, and pleural fluid of adult patients who underwent diagnostic thoracentesis between October 2013 and December 2018. Pleural effusions were classified into the following five categories: transudative, malignant, parapneumonic, tuberculous, and other. The performance of five different classifiers, including multinomial logistic regression, support vector machine, random forest, extreme gradient boosting, and light gradient boosting machine (LGB), was evaluated in terms of accuracy and area under the receiver operating characteristic curve through fivefold cross-validation. Hybrid feature selection was applied to determine the most relevant features for classifying pleural effusion. Results: We analyzed 2,253 patients (training set, n = 1,459; validation set, n = 365; extra-validation set, n = 429) and found that the LGB model achieved the best performance in both validation and extra-validation sets. After feature selection, the accuracy of the LGB model with the selected 18 features was equivalent to that with all 49 features (mean ± standard deviation): 0.818 ± 0.012 and 0.777 ± 0.007 in the validation and extra-validation sets, respectively. The model's mean area under the receiver operating characteristic curve was as high as 0.930 ± 0.042 and 0.916 ± 0.044 in the validation and extra-validation sets, respectively. In our model, pleural lactate dehydrogenase, protein, and adenosine deaminase levels were the most important factors for classifying pleural effusions. Conclusions: Our LGB model showed satisfactory performance for differential diagnosis of the common causes of pleural effusions. This model could provide clinicians with valuable information regarding the major differential diagnoses of pleural diseases.

Machine learning model for predicting immediate postoperative desaturation using spirometry signal data.

Postoperative desaturation is a common post-surgery pulmonary complication. The real-time prediction of postoperative desaturation can become a preventive measure, and real-time changes in spirometry data can provide valuable information on respiratory mechanics. However, there is a lack of related research, specifically on using spirometry signals as inputs to machine learning (ML) models. We developed an ML model and postoperative desaturation prediction index (DPI) by analyzing intraoperative spirometry signals in patients undergoing laparoscopic surgery. We analyzed spirometry data from patients who underwent laparoscopic, robot-assisted gynecologic, or urologic surgery, identifying postoperative desaturation as a peripheral arterial oxygen saturation level below 95%, despite facial oxygen mask usage. We fitted the ML model on two separate datasets collected during different periods. (Datasets A and B). Dataset A (Normal 133, Desaturation 74) was used for the entire experimental process, including ML model fitting, statistical analysis, and DPI determination. Dataset B (Normal 20, Desaturation 4) was only used for verify the ML model and DPI. Four feature categories-signal property, inter-/intra-position correlation, peak value/interval variability, and demographics-were incorporated into the ML models via filter and wrapper feature selection methods. In experiments, the ML model achieved an adequate predictive capacity for postoperative desaturation, and the performance of the DPI was unbiased.

Preliminary analysis of predicting the first recurrence in patients with neovascular age-related macular degeneration using deep learning.

BACKGROUND: To predict, using deep learning, the first recurrence in patients with neovascular age-related macular degeneration (nAMD) after three monthly loading injections of intravitreal anti-vascular endothelial growth factor (anti-VEGF). METHODS: Optical coherence tomography (OCT) images were obtained at baseline and after the loading phase. The first recurrence was defined as the initial appearance of a new retinal hemorrhage or intra/subretinal fluid accumulation after the initial resolution of exudative changes after three loading injections. Standard U-Net architecture was used to identify the three retinal fluid compartments, which include pigment epithelial detachment, subretinal fluid, and intraretinal fluid. To predict the first recurrence of nAMD, classification learning was conducted to determine whether the first recurrence occurred within three months after the loading phase. The recurrence classification architecture was built using ResNet50. The model with retinal regions of interest of the entire region and fluid region on OCT at baseline and after the loading phase is presented. RESULTS: A total of 1,444 eyes of 1,302 patients were included. The mean duration until the first recurrence after the loading phase was 8.20 ± 15.56 months. The recurrence classification system revealed that the model with the fluid region of OCT after the loading phase provided the highest classification performance, with an area under the receiver operating characteristic curve (AUC) of 0.725 ± 0.012. Heatmap analysis revealed that three pathological fluids, subsided choroidal neovascularization lesions, and hyperreflective foci were important areas for the first recurrence. CONCLUSIONS: The deep learning algorithm allowed for the prediction of the first recurrence for three months after the loading phase with adequate feasibility. An automated prediction system may assist in establishing patient-specific treatment plans and the provision of individualized medical care for patients with nAMD.

Using spectral and temporal filters with EEG signal to predict the temporal lobe epilepsy outcome after antiseizure medication via machine learning.

Epilepsy is a neurological disorder in which the brain is transiently altered. Predicting outcomes in epilepsy is essential for providing feedback that can foster improved outcomes in the future. This study aimed to investigate whether applying spectral and temporal filters to resting-state electroencephalography (EEG) signals could improve the prediction of outcomes for patients taking antiseizure medication to treat temporal lobe epilepsy (TLE). We collected EEG data from a total of 46 patients (divided into a seizure-free group (SF, n = 22) and a non-seizure-free group (NSF, n = 24)) with TLE and retrospectively reviewed their clinical data. We segmented spectral and temporal ranges with various time-domain features (Hjorth parameters, statistical parameters, energy, zero-crossing rate, inter-channel correlation, inter-channel phase locking value and spectral information derived from Fourier transform, Stockwell transform, and wavelet transform) and compared their performance by applying an optimal frequency strategy, an optimal duration strategy, and a combination strategy. For all time-domain features, the optimal frequency and time combination strategy showed the highest performance in distinguishing SF patients from NSF patients (area under the curve (AUC) = 0.790 ± 0.159). Furthermore, optimal performance was achieved by utilizing a feature vector derived from statistical parameters within the 39- to 41-Hz frequency band with a window length of 210 s, as evidenced by an AUC of 0.748. By identifying the optimal parameters, we improved the performance of the prediction model. These parameters can serve as standard parameters for predicting outcomes based on resting-state EEG signals.