Aids Clinicians Research Articles

Prognostic value of multi-PLD ASL radiomics in acute ischemic stroke

IntroductionEarly prognosis prediction of acute ischemic stroke (AIS) can support clinicians in choosing personalized treatment plans. The aim of this study is to develop a machine learning (ML) model that uses multiple post-labeling delay times (multi-PLD) arterial spin labeling (ASL) radiomics features to achieve early and precise prediction of AIS prognosis.MethodsThis study enrolled 102 AIS patients admitted between December 2020 and September 2024. Clinical data, such as age and baseline National Institutes of Health Stroke Scale (NIHSS) score, were collected. Radiomics features were extracted from cerebral blood flow (CBF) images acquired through multi-PLD ASL. Features were selected using least absolute shrinkage and selection operator regression, and three models were developed: a clinical model, a CBF radiomics model, and a combined model, employing eight ML algorithms. Model performance was assessed using receiver operating characteristic curves and decision curve analysis (DCA). Shapley Additive exPlanations was applied to interpret feature contributions.ResultsThe combined model of extreme gradient boosting demonstrated superior predictive performance, achieving an area under the curve (AUC) of 0.876. Statistical analysis using the DeLong test revealed its significant outperformance compared to both the clinical model (AUC = 0.658, p < 0.001) and the CBF radiomics model (AUC = 0.755, p = 0.002). The robustness of all models was confirmed through permutation testing. Furthermore, DCA underscored the clinical utility of the combined model. The prognostic prediction of AIS was notably influenced by the baseline NIHSS score, age, as well as texture and shape features of CBF.ConclusionThe integration of clinical data and multi-PLD ASL radiomics features in a model offers a secure and dependable approach for predicting the prognosis of AIS, particularly beneficial for patients with contraindications to contrast agents. This model aids clinicians in devising individualized treatment plans, ultimately enhancing patient prognosis.

Machine learning-based predictive models for perioperative major adverse cardiovascular events in patients with stable coronary artery disease undergoing noncardiac surgery.

Accurate prediction of perioperative major adverse cardiovascular events (MACEs) is crucial, as it not only aids clinicians in comprehensively assessing patients' surgical risks and tailoring personalized surgical and perioperative management plans, but also for information-based shared decision-making with patients and efficient allocation of medical resources. This study developed and validated a machine learning (ML) model using accessible preoperative clinical data to predict perioperative MACEs in stable coronary artery disease (SCAD) patients undergoing noncardiac surgery (NCS). We collected data from 9171 adult SCAD patients who underwent NCS and extracted 64 preoperative variables. First, the optimal data imputation, resampling, and feature selection methods were compared and selected to deal with missing data values and imbalances. Then, nine independent machine learning models (logistic regression (LR), support vector machine, Gaussian Naive Bayes (GNB), random forest, gradient boosting decision tree (GBDT), extreme gradient boosting (XGBoost), light gradient boosting machine, categorical boosting (CatBoost), and deep neural network) and a stacking ensemble model were constructed and compared with the validated Revised Cardiac Risk Index's (RCRI) model for predictive performance, which was evaluated using the area under the receiver operating characteristic curve (AUROC), the area under the precision-recall curve (AUPRC), calibration curve, and decision curve analysis (DCA). To reduce overfitting and enhance robustness, we performed hyperparameter tuning and 5-fold cross-validation. Finally, the Shapley additive interpretation (SHAP) method and a partial dependence plot (PDP) were used to determine the optimal ML model. Of the 9,171 patients, 514 (5.6 %) developed MACEs. 24 significant preoperative features were selected for model development and evaluation. All ML models performed well, with AUROC above 0.88 and AUPRC above 0.39, outperforming the AUROC (0.716) and AUPRC (0.185) of RCRI (P < 0.001). The best independent model was XGBoost (AUROC = 0.898, AUPRC = 0.479). The calibration curve accurately predicted the risk of MACEs (Brier score = 0.040), and the DCA results showed that XGBoost had a high net benefit for predicting MACEs. The top-ranked stacking ensemble model, consisting of CatBoost, GBDT, GNB, and LR, proved to be the best (AUROC 0.894, AUPRC 0.485). We identified the top 20 most important features using the mean absolute SHAP values and depicted their effects on model predictions using PDP. This study combined missing-value imputation, feature screening, unbalanced data processing, and advanced machine learning methods to successfully develop and verify the first ML-based perioperative MACEs prediction model for patients with SCAD, which is more accurate than RCRI and enables effective identification of high-risk patients and implementation of targeted interventions to reduce the incidence of MACEs.

Predictive model for PSA persistence after radical prostatectomy using machine learning algorithms.

To evaluate the efficacy of a machine learning model for predicting prostate-specific antigen (PSA) persistence after radical prostatectomy (RP). Data from 470 patients who underwent RP at the Affiliated Hospital of Qingdao University from January 2018 to June 2021 were retrospectively analyzed. Ten risk factors, including age, body mass index (BMI), preoperative PSA, biopsy Gleason score, total prostate specific antigen density (PSAD), clinical tumor stage, clinical lymph node status, seminal vesicle invasion, capsular invasion and positive surgical margin, were included in the analysis. The data were randomly divided into a training set and a test set at a ratio of 7:3, and seven different machine learning algorithms were compared. The confusion matrix, receiver operating characteristic (ROC) curve and area under the ROC curve (AUC) were used to evaluate the diagnostic performance of the model, and the random forest algorithm found to be the optimal prediction model. In the entire cohort, 142 (30.21%) patients developed PSA persistence. Based on all included risk factors, the random forest model had the best effect among the seven models, with an AUC of 0.8607 in the training set and 0.8011 in the test set. The feature importance results showed that capsular invasion, positive surgical margin, preoperative PSA and biopsy Gleason score were the four most important risk factors for PSA persistence after RP. The Random Forest algorithm performed excellently in this study and can be used to construct a predictive model for PSA persistence. By incorporating clinical data from the Asian region and exploring the risk factors for PSA persistence, this study contributes to the existing research and aids clinicians in assessing the risk of PSA persistence occurrence, enabling timely treatment planning and improving patient prognosis.

Task-Specific Transformer-Based Language Models in Health Care: Scoping Review.

Transformer-based language models have shown great potential to revolutionize health care by advancing clinical decision support, patient interaction, and disease prediction. However, despite their rapid development, the implementation of transformer-based language models in health care settings remains limited. This is partly due to the lack of a comprehensive review, which hinders a systematic understanding of their applications and limitations. Without clear guidelines and consolidated information, both researchers and physicians face difficulties in using these models effectively, resulting in inefficient research efforts and slow integration into clinical workflows. This scoping review addresses this gap by examining studies on medical transformer-based language models and categorizing them into 6 tasks: dialogue generation, question answering, summarization, text classification, sentiment analysis, and named entity recognition. We conducted a scoping review following the Cochrane scoping review protocol. A comprehensive literature search was performed across databases, including Google Scholar and PubMed, covering publications from January 2017 to September 2024. Studies involving transformer-derived models in medical tasks were included. Data were categorized into 6 key tasks. Our key findings revealed both advancements and critical challenges in applying transformer-based models to health care tasks. For example, models like MedPIR involving dialogue generation show promise but face privacy and ethical concerns, while question-answering models like BioBERT improve accuracy but struggle with the complexity of medical terminology. The BioBERTSum summarization model aids clinicians by condensing medical texts but needs better handling of long sequences. This review attempted to provide a consolidated understanding of the role of transformer-based language models in health care and to guide future research directions. By addressing current challenges and exploring the potential for real-world applications, we envision significant improvements in health care informatics. Addressing the identified challenges and implementing proposed solutions can enable transformer-based language models to significantly improve health care delivery and patient outcomes. Our review provides valuable insights for future research and practical applications, setting the stage for transformative advancements in medical informatics.

BIOM-47. SINGLE NUCLEOTIDE POLYMORPHISM IN SMARCB1 MUTATION FOR SCHWANNOMATOSIS

Abstract Schwannomatosis, a form of neurofibromatosis, involves critical roles for the SMARCB1 gene. Clinical diagnosis guidelines recommend considering pathogenic variants or suspecting schwannomatosis to exclude splicing-related intron mutations. However, some patients may initially present with non-significant mutations and lack clinical signs of schwannoma, leading to misdiagnosis, missed annual assessments, malignancy awareness, and genetic consultation opportunities during pregnancy. Here, we reported a patient with a sporadic synonymous mutation in the SMARCB1 gene (SNP c.1032 C&amp;gt;T (p.Gly344Gly(GGC&amp;gt;GGT)) in exon 8), not previously associated with schwannomatosis. This patient, a 55-year-old female with an 8-year history of schwannomatosis, presented to the neurosurgical department for recurrent tumor removal. Initially, the clinical impact of the SMARCB1 mutation was undetermined, and other genes tested were wild-type. Both schwannomas exhibited a mosaic loss of nuclear SMARCB1 protein, ranging from 10% to 60%. Protein assays confirmed low levels of SMARCB1 protein, with a significant loss in the distal thigh schwannoma. It highlights the tumorigenic potential of single nucleotide polymorphism in the SMARCB1 gene and aids clinicians in making more accurate diagnoses in patients with pathologically confirmed schwannoma. It underscores the importance of long-term follow-up, increased awareness of recurrence and malignancy, and timely surgical planning in such patients.

BrainSim: AI-Driven Brain MRI Simulation for Alzheimer’s Disease Diagnosis

Alzheimer's Disease, a progressive neurodegenerative disorder with no cure, presents a formidable challenge to global healthcare. The simulation of brain states, encompassing both rejuvenation and aging, aids clinicians in understanding disease progression and early detection. Recent advances in deep adversarial neural networks, effective in generating age-varied facial images, are now applied to brain MRI simulations across different time states. However, existing methods for brain MRI synthesis predominantly focus on aging simulation, lacking rejuvenation capabilities, and relying heavily on longitudinal data. Access to rejuvenated brain scans is essential as it can help researchers track the structural changes and biomarkers in the early stage of AD, facilitating AD research and personalized treatments. Additionally, previous methods are often 2D-based, failing to capture complex 3D spatial information obtained by 3D scans. To overcome these limitations, I propose BrainSim, a novel longitudinal brain MRI synthesis framework. Specifically, BrainSim leverages a bidirectional cycle-consistent generative neural network to generate rejuvenated and aged MRI scans simultaneously using cross-sectional scans only. BrainSim incorporates a 3D conditional U-Net as its fundamental model, facilitating the direct generation of 3D MRI scans conditioned on the subject's health state and sex. BrainSim was evaluated against benchmark models using longitudinal data and a pre-trained age predictor to assess the quality of generated images. Experiments using the ADNI dataset show BrainSim's state-of-the-art performance in brain aging and rejuvenation tasks, marking significant advancements in MRI synthesis for Alzheimer's research and drug development.

Proteomic analysis of post-COVID condition: Insights from plasma and pellet blood fractions

BackgroundPersistent symptoms extending beyond the acute phase of SARS-CoV-2 infection, known as Post-COVID condition (PCC), continue to impact many individuals years after the COVID-19 pandemic began. This highlights an urgent need for a deeper understanding and effective treatments. While significant progress has been made in understanding the acute phase of COVID-19 through omics-based approaches, the proteomic alterations linked to the long-term effects of the infection remain underexplored. This study aims to investigate these proteomic changes and develop a method for stratifying disease severity. MethodsUsing Sequential Window Acquisition of All Theoretical Fragment Ion Mass Spectra (SWATH-MS) technology, we performed comprehensive proteomic profiling of blood samples from 65 PCC patients. Both plasma and pellet (cellular components) fractions were analyzed to capture a wide array of proteomic changes associated with PCC. ResultsProteomic profiling revealed distinct differences between symptomatic and asymptomatic PCC patients. In the plasma fraction, symptomatic patients exhibited significant upregulation of proteins involved in coagulation, immune response, oxidative stress, and various metabolic processes, while certain immunoglobulins and proteins involved in cellular stress responses were downregulated. In the pellet fraction, symptomatic patients showed upregulation of proteins related to immune response, coagulation, oxidative stress, and metabolic enzymes, with downregulation observed in components of the complement system, glycolysis enzymes, and cytoskeletal proteins. A key outcome was the development of a novel severity scale based on the concentration of identified proteins, which correlated strongly with the clinical symptoms of PCC. This scale, derived from unsupervised clustering analysis, provides precise quantification of PCC severity, enabling effective patient stratification. ConclusionsThe identified proteomic alterations offer valuable insights into the molecular mechanisms underlying PCC, highlighting potential biomarkers and therapeutic targets. This research supports the development of tailored clinical interventions to alleviate persistent symptoms, ultimately enhancing patient outcomes and quality of life. The quantifiable measure of disease severity aids clinicians in understanding the condition in individual patients, facilitating personalized treatment plans and accurate monitoring of disease progression and response to therapy.

Early prediction of platelet recovery with immature platelet fraction in patients receiving hematopoietic stem cell transplantation.

Hematopoietic stem cell transplantation (HSCT) is pivotal in treating hematologic disorders, yet it poses the risk of post-transplantation pancytopenia. Prophylactic platelet transfusions are often administered to mitigate this risk. Utilizing practical markers, such as immature platelet fraction (IPF), to predict hematopoietic recovery in advance could reduce unnecessary prophylactic transfusions. Our prospective study, involving 53 HSCT patients at Taipei Veterans General Hospital between September 2022 and May 2023, utilized the Sysmex XN analyzer to assess peripheral blood cell parameters. We investigated whether IPF could predict platelet recovery early, determined the optimal cut-off value, and compared platelet usage. Neutrophil and platelet engraftment occurred 10 (median; range: 10-12) and 15 (median; range: 15-18) days post-HSCT. Notably, 71.7% of patients exhibited an IPF increase exceeding 2% before platelet recovery. The optimal cut-off IPF on day 10 for predicting platelet recovery within five days was 2.15% (specificity 0.89, sensitivity 0.65). On average, patients received 3.89 units of post-transplantation platelet transfusion. Our results indicate that IPF serves as a predictive marker for platelet engraftment, peaking before the increase in platelet count. This insight aids clinicians in assessing the need for prophylactic platelet transfusions. Integrating reference IPF values alongside platelet counts enhances the accuracy of evaluating a patient's hematopoietic recovery status. Anticipating the timing of platelet recovery optimizes blood product usage and mitigates transfusion reaction risks.

A dynamic online nomogram for predicts delayed postoperative bleeding after colorectal polyp surgery

This study aims to analyze the risk factors associated with delayed postoperative bleeding (DPPB) following colorectal polyp surgery, develop a dynamic nomogram and evaluate the model efficacy, provide a reference for clinicians to identify the patients at high risk of DPPB. Retrospective study was done on patients who underwent endoscopic colorectal polypectomy at the First Hospital of Lanzhou University from January 2020 to March 2023. Differences between the group with and without DPPB were compared, and independent risk factors for DPPB occurrence were identified through univariate analysis and combination LASSO and logistic regression. A dynamic nomogram was constructed based on multiple logistic regression to predict DPPB following colorectal polyp surgery. Model evaluation included receiver operating characteristic (ROC), Calibration curve, Decision curve analysis (DCA). DPPB occurred in 38 of the 1544 patients included. multivariate analysis showed that direct oral anticoagulants (DOACs), polyp location in the right hemi colon, polyp diameter, drink, and prophylactic hemoclips were the independent risk factors for DPPB and dynamic nomogram were established. Model validation indicated area under the ROC curve values of 0.936, 0.796, and 0.865 for the training set, validation set, and full set, respectively. The calibration curve demonstrated a strong alignment between the predictions of the column-line diagram model and actual observations. The decision curve analysis (DCA) displayed a significant net clinical benefit across the threshold probability range of 0–100%. The dynamic nomogram aids clinicians in identifying high-risk patients, enabling personalized diagnosis and treatment.

Comparing Graph Sample and Aggregation (SAGE) and Graph Attention Networks in the Prediction of Drug-Gene Associations of Extended-Spectrum Beta-Lactamases in Periodontal Infections and Resistance.

Gram-negative bacteria exhibit more antibiotic resistance than gram-positive bacteria due to their cell wall structure and composition differences. Porins, or protein channels in these bacteria, can allow small, hydrophilic antibiotics to diffuse, affecting their susceptibility. Mutations in porin protein genes can also impair antibiotic entry. Predicting drug-gene associations of extended-spectrum beta-lactamases (ESBLs) is crucial as they confer resistance to beta-lactam antibiotics, challenging the treatment of infections. This aids clinicians in selecting suitable treatments, optimizing drug usage, enhancing patient outcomes, and controlling antibiotic resistance in healthcare settings. Graph-based neural networks can predict drug-gene associations in periodontal infections and resistance. The aim of the study was to predict drug-gene associations of ESBLs in periodontal infections and resistance. The study focuses on analyzing drug-gene associations using probes and drugs. The data was converted into graph language, assigning nodes and edges for drugs and genes. Graph neural networks (GNNs) and similar algorithms were implemented using Google Colab and Python. Cytoscape and CytoHubba are open-source software platforms used for network analysis and visualization. GNNswere used for tasks like node classification, link prediction, and graph-level prediction. Three graph-based models were used: graph convolutional network (GCN), Graph SAGE, and graph attention network (GAT). Each model was trained for 200 epochs using the Adam optimizer with a learning rate of 0.01 and a weight decay of 5e-4. The drug-gene association network has 57 nodes, 79 edges, and a 2.730 characteristic path length. Its structure, organization, and connectivity are analyzed using the GCN and Graph SAGE, which show high accuracy, precision, recall, and an F1-score of 0.94. GAT's performance metrics are lower, with an accuracy of 0.68, precision of 0.47, recall of 0.68, and F1-score of 0.56, suggesting that it may not be as effective in capturing drug-gene relationships. Compared to ESBLs, both GCN and Graph SAGE demonstrate excellent performance with accuracy, precision, recall, and an F1-score of 0.94. These results indicate that GCN and Graph SAGE are highly effective in predicting drug-gene associations related to ESBLs. GCN and Graph SAGE outperform GAT in predicting drug-gene associations for ESBLs. Improvements include data augmentation, regularization, and cross-validation. Ethical considerations, fairness, and open-source implementations are crucial for future research in precision periodontal treatment.

Predictive models for personalized precision medical intervention in spontaneous regression stages of cervical precancerous lesions

BackgroundDuring the prolonged period from Human Papillomavirus (HPV) infection to cervical cancer development, Low-Grade Squamous Intraepithelial Lesion (LSIL) stage provides a critical opportunity for cervical cancer prevention, giving the high potential for reversal in this stage. However, there is few research and a lack of clear guidelines on appropriate intervention strategies at this stage, underscoring the need for real-time prognostic predictions and personalized treatments to promote lesion reversal.MethodsWe have established a prospective cohort. Since 2018, we have been collecting clinical data and pathological images of HPV-infected patients, followed by tracking the progression of their cervical lesions. In constructing our predictive models, we applied logistic regression and six machine learning models, evaluating each model’s predictive performance using metrics such as the Area Under the Curve (AUC). We also employed the SHAP method for interpretative analysis of the prediction results. Additionally, the model identifies key factors influencing the progression of the lesions.ResultsModel comparisons highlighted the superior performance of Random Forests (RF) and Support Vector Machines (SVM), both in clinical parameter and pathological image-based predictions. Notably, the RF model, which integrates pathological images and clinical multi-parameters, achieved the highest AUC of 0.866. Another significant finding was the substantial impact of sleep quality on the spontaneous clearance of HPV and regression of LSIL.ConclusionsIn contrast to current cervical cancer prediction models, our model’s prognostic capabilities extend to the spontaneous regression stage of cervical cancer. This model aids clinicians in real-time monitoring of lesions and in developing personalized treatment or follow-up plans by assessing individual risk factors, thus fostering lesion spontaneous reversal and aiding in cervical cancer prevention and reduction.Graphical

Detailed regimens for the prolonged β-lactam infusion therapy

A recent systematic review and meta-analysis of randomized controlled trials (RCTs) evaluated the efficacy and safety of prolonged versus intermittent β-lactam infusion in adult sepsis patients. The findings revealed a significant decrease in all-cause mortality and marked clinical success in the prolonged infusion group. Unfortunately, however, the manuscript lacked data and discussion for the specific regimens of prolonged β-lactam infusion defined in the included 15 RCT studies, which are herein additionally provided. Excluding one RCT, all protocols adopted a continuous infusion for the prolonged treatment. Except for three RCTs, dosages and timings of bolus injection were clearly defined. The total daily antibiotic dose for the continuous therapy was equivalent to those recommended for intermittent therapy. We believe this supplementary data aids clinicians in providing prolonged β-lactam infusions, contributing to enhanced treatment outcomes for patients suffering from severe sepsis or septic shock.

P-246 Use of image recognition to quantify oocyte quality improved the accuracy of prediction of blastulation by 26-48%

Abstract Study question Can an AI oocyte quality model improve blastulation prediction accuracy? Summary answer An AI oocyte quality model presented an overall accuracy of 61-74%. Exhibiting an improvement of 26-48% for prediction of blastulation over random. What is known already Nowadays, the assessment of whether sufficient oocytes have been cryopreserved is purely based on patient age and number of oocytes collected. No morphological abnormalities provide quantitative guidance about the reproductive potential of the oocyte. Additionally, interobserver agreement and prediction of blastulation by experienced embryologists has been reported to be worse than chance (Behr et al., 2018). Therefore, there is a need to implement oocyte quality assessment tools to predict blastulation in oocyte donation, cryopreservation programs and IVF treatments. Study design, size, duration A retrospective study using an AI oocyte quality score (CHLOE-OQ, Fairtility Ltd) was performed in three private clinics evaluating 1749 oocytes that were inseminated with ICSI and cultured between October 2019 and December 2021. Overall and per clinic analysis were conducted (Clinic 1 n = 1131; Clinic 2 n = 576; Clinic 3: n = 42). Participants/materials, setting, methods An AI model provided an oocyte quality (OQ) score to assess prediction of blastulation, measured by binary logistic regression (AUC) per clinic and overall. The correlation between the OQ score and blastulation was measured using t-test. OQ score was classified into groups, Group A: 0.9-1.0, Group B: 0.7-0.9, Group C: 0.4-0.7 and Group D: 0-0.4. Blastulation rate was assessed in each OQ score subgroup using chi-square. Main results and the role of chance OQ score prediction of blastulation presented an overall accuracy of 63% (1107/1749). Accuracy per clinic ranged from 61-74%, Clinic 1: 61% (690/1131), Clinic 2: 67% (386/576) and Clinic 3: 74% (31/42). Exhibiting an improvement of 24-48% for prediction of blastulation over random. Oocyte quality score was predictive of blastulation per clinic and overall Clinic 1: AUC 0.62, n = 1130; Clinic 2: AUC 0.62, n = 575; Clinic 3: AUC 0.78, n = 41; Overall: AUC=0.62, p &amp;lt; 0.001, n = 1749). Overall blastocyst development rate was 53% (929/1749). OQ score was higher in oocytes that blastulated than those that did not blastulate (0.85 ± 0.14 vs 0.74 ± 0.2, p &amp;lt; 0.001). There was a significant direct association between OQ score subgroups and blastulation rate. Group A with the highest score subgroup had the highest blastulation rate [61% (554/903), p &amp;lt; 0.05]. The blastulation rates for Group B [53% (283/539) p &amp;lt; 0.001], Group C [38%, (71/188), p &amp;lt; 0.001] and Group D [18% (21/119), p &amp;lt; 0.001] were decreasingly lower. Limitations, reasons for caution This study was done in 3 centers in the same country. Therefore, further studies should focus on diversifying in terms of geographical location to address generalization of this AI assessment. Further studies should focus on comparing the efficacy of blastulation prediction by humans. Wider implications of the findings Using AI to do an objective oocyte assessment aids clinicians in determining the optimal time to transition to oocyte donor programs. It assists the decision-making for oocyte banking, minimizing unnecessary cycles and guiding timely additional oocyte collections. This approach at the point of cryopreservation, mitigates risks associated with delayed decisions. Trial registration number N/A

O-321 A foundation model for embryology trained on time-lapse images

Abstract Study question Can a generalizable computer vision foundation model be trained using self-supervised learning on time-lapse embryo images to perform multiple clinically relevant downstream tasks? Summary answer We developed FEMI, a foundation model trained on eight million time-lapse images, to perform multiple clinical tasks, including blastocyst quality scoring, ploidy prediction, and segmentation. What is known already In vitro fertilization success critically depends on choosing viable embryos, a process hampered by current limited diagnostic tools, high costs, and ethical concerns. In recent years, various medical fields have increasingly explored foundation models using vision transformer architectures. These models, trained in a self-supervised manner on vast unlabeled image datasets, perform various clinically relevant tasks. Once trained on a massive unlabeled dataset, the encoder portion of the foundation model can be extracted and subsequently fine-tuned with a labeled dataset to perform various clinically relevant tasks like classification, regression, and segmentation. Study design, size, duration The foundation model was trained on 8 million Embryoscope® (E-SD) and Embryoscope+® (E+) images spanning clinics in the United States, Canada, and Europe. Datasets contained various additional information, such as ploidy status, blastocyst scores (numerical values from 3 to 14 based on Zhan et al. 2020), and segmentation masks (for trophectoderm [TE], inner cell mass [ICM], and zona pellucida [ZP]) that were used for training downstream tasks. Participants/materials, setting, methods A masked autoencoder architecture is trained on time-lapse images to create a foundation model (FEMI). The encoder from the autoencoder is extracted and fine-tuned on three image-based downstream tasks: ploidy prediction, blastocyst scoring, and embryo component segmentation. The performance of the fine-tuned foundation model is compared to VGG16 architectures, specifically trained for each downstream task. We evaluated performances using the area under the receiver-operating-characteristic (AUROC), mean absolute error (MAE), and mean intersection over union (mIoU). Main results and the role of chance The training dataset for the downstream tasks consisted of image data and labels, including PGT-A ploidy results (euploid or aneuploid), blastocyst scores (3-14), and segmentation masks. No clinical data like maternal age was used in order to assess FEMI’s learning capabilities from only time-lapse images against VGG16 model baselines. Both FEMI and baseline models used the same data splits for consistent comparisons. For ploidy prediction, models trained on a Weill Cornell Medicine (WCM) dataset were validated using WCM, Florida, and Spain data. FEMI achieved a 0.610 ± 0.004 AUROC, outperforming the baseline’s 0.590 ± 0.005. In blastocyst score prediction, FEMI attained a superior MAE of 0.099 ± 0.002 compared to the baseline’s 0.118 ± 0.001. For embryo segmentation (TE, ICM, ZP), a publicly available dataset from Simon Fraser University was split for training and validation. FEMI exceeded the baseline in mIoU for TE segmentation (0.738 ± 0.002 vs. 0.726 ± 0.002) and was comparable in ICM (0.832 ± 0.008 vs. 0.821 ± 0.011) and ZP segmentation (0.779 ± 0.002 vs. 0.778 ± 0.010). Limitations, reasons for caution The current version of FEMI is trained on a subset of our available datasets and can be improved through further training. Moreover, the downstream ploidy and blastocyst score models may be biased by errors in PGT results and the subjectivity of embryologists, respectively. Wider implications of the findings A generalizable foundational model for IVF time-lapse imaging aids clinicians in embryo selection by providing various clinical insights for better decision-making. FEMI, once fully trained, will be publicly accessible to the scientific community allowing for researchers to fine-tune FEMI on their own clinic-specific applications with their own image datasets. Trial registration number not applicable

Segmentation model of soft tissue sarcoma based on self-supervised learning.

Soft tissue sarcomas, similar in incidence to cervical and esophageal cancers, arise from various soft tissues like smooth muscle, fat, and fibrous tissue. Effective segmentation of sarcomas in imaging is crucial for accurate diagnosis. This study collected multi-modal MRI images from 45 patients with thigh soft tissue sarcoma, totaling 8,640 images. These images were annotated by clinicians to delineate the sarcoma regions, creating a comprehensive dataset. We developed a novel segmentation model based on the UNet framework, enhanced with residual networks and attention mechanisms for improved modality-specific information extraction. Additionally, self-supervised learning strategies were employed to optimize feature extraction capabilities of the encoders. The new model demonstrated superior segmentation performance when using multi-modal MRI images compared to single-modal inputs. The effectiveness of the model in utilizing the created dataset was validated through various experimental setups, confirming the enhanced ability to characterize tumor regions across different modalities. The integration of multi-modal MRI images and advanced machine learning techniques in our model significantly improves the segmentation of soft tissue sarcomas in thigh imaging. This advancement aids clinicians in better diagnosing and understanding the patient's condition, leveraging the strengths of different imaging modalities. Further studies could explore the application of these techniques to other types of soft tissue sarcomas and additional anatomical sites.

What influences post-operative opioid requirements for tibial fractures?

Currently there are few opioid prescribing guidelines for orthopaedic fractures. Long-term post-surgical analgesia requirements, understandably, vary between orthopaedic cases. Our study aims to provide detailed information to clinicians and policy makers, on the opioid requirement associations for patients sustaining tibial fractures. This study reviewed all patients sustaining an isolated tibial fracture at a major trauma centre that were operated on within 1month of injury, from 2015 to 2022. The total opioid dosage used each month in morphine milligrams equivalents (MME) and the number of days opioids were used each month, within the first-year post-surgery were collected, representing the strength and coverage of opioid analgesia in the post-operative stage, respectively. We compared opioid strength and coverage requirements with types of definitive fracture fixations, location, fracture type and concurrent patient medical comorbidities to assess for any trends. A total of 1814 patients sustaining a combined of 1970 fractures were included in the study. Tibial plateau fractures had the highest opioid strength and coverage requirements in each month and the entire year (p < .05). Across all fracture locations, Ex Fix frame showed higher opioid strength and coverage requirements compared to both IM nailing and plate ORIF. With regard to opioid coverage in the presence of specific comorbidities, only chronic kidney disease (quotient: 1.37, 95% Confidence interval [95%CI] = 1.19-1.55, p = .002) and hypertension (quotient: 1.34, 95%CI = 1.14-1.53, p = .009) showed significance at the 1-year overall level. For opioid strength, Chronic Kidney Disease (quotient: 1.72, 95%CI = 1.41-2.03 p = .005) and COPD (quotient: 1.90, 95%CI = 1.44-2.36, p = .014), show significance at the 1-year overall level. Our study details opioid requirements post-surgery amongst tibial fractures with subgroup analysis assessing opioid needs amongst specific fracture locations, types, surgical techniques and medical comorbidities. This framework aids clinicians in anticipating rehabilitation and assists in risk stratifying patients at injury onset.

Artificial intelligence model for tumoral clinical decision support systems

Background and ObjectiveComparative diagnostic in brain tumor evaluation makes possible to use the available information of a medical center to compare similar cases when a new patient is evaluated. By leveraging Artificial Intelligence models, the proposed system is able of retrieving the most similar cases of brain tumors for a given query. The primary objective is to enhance the diagnostic process by generating more accurate representations of medical images, with a particular focus on patient-specific normal features and pathologies. A key distinction from previous models lies in its ability to produce enriched image descriptors solely from binary information, eliminating the need for costly and difficult to obtain tumor segmentation. MethodsThe proposed model uses Artificial Intelligence to detect patient features to recommend the most similar cases from a database. The system not only suggests similar cases but also balances the representation of healthy and abnormal features in its design. This not only encourages the generalization of its use but also aids clinicians in their decision-making processes. This generalization makes possible for future research in different medical diagnosis areas with almost not any change in the system. ResultsWe conducted a comparative analysis of our approach in relation to similar studies. The proposed architecture obtains a Dice coefficient of 0.474 in both tumoral and healthy regions of the patients, which outperforms previous literature. Our proposed model excels at extracting and combining anatomical and pathological features from brain Magnetic Resonances (MRs), achieving state-of-the-art results while relying on less expensive label information. This substantially reduces the overall cost of the training process. Our findings highlight the significant potential for improving the efficiency and accuracy of comparative diagnostics and the treatment of tumoral pathologies. ConclusionsThis paper provides substantial grounds for further exploration of the broader applicability and optimization of the proposed architecture to enhance clinical decision-making. The novel approach presented in this work marks a significant advancement in the field of medical diagnosis, particularly in the context of Artificial Intelligence-assisted image retrieval, and promises to reduce costs and improve the quality of patient care using Artificial Intelligence as a support tool instead of a black box system.

Investigation of the Internal Structure and Radiological Characteristics of Distant Metastases to the Jaws: A Retrospective Study

Aim: Distant metastases to the jaws are uncommon but carry a grave prognosis, with an average survival of only a few months. This study aimed to investigate the internal structure and radiological features of metastatic lesions in the jaws, as well as their clinical presentations, to better understand their diagnostic characteristics and guide appropriate management. Materials and Methods: A retrospective review was conducted using the patient database of the Department of Oral and Maxillofacial Radiology at Istanbul University. Among 4,908 patients who underwent cone beam computed tomography (CBCT), 14 cases with histopathologically confirmed metastatic tumors to the jaws were identified. Clinical data, imaging modalities (orthopantomography, CBCT, magnetic resonance imaging, ultrasound), and histopathological findings were reviewed. The primary tumor sites, radiographic appearances (osteolytic vs. osteoblastic lesions), and histopathological patterns were examined. Results: Thirteen lesions were located in the mandible and one in the maxilla. The most common primary tumors originated from the breast, prostate, lung, kidney, and colon. While breast and prostate metastases appeared mostly as sclerotic (radiopaque) lesions, lung, kidney, and colon metastases typically presented as osteolytic (radiolucent) defects. In some cases, pathologic fractures and cortical destruction were evident. Histopathological evaluation confirmed metastatic carcinoma in all cases, demonstrating characteristic osteoblastic activity in breast and prostate cancers. Conclusion: Recognizing the radiological and histopathological features of distant metastases to the jaws is crucial for prompt diagnosis, accurate identification of the primary tumor, and appropriate treatment planning. Awareness of these lesions aids clinicians in differentiating metastases from other jaw pathologies and underscores the importance of multidisciplinary collaboration.

A clinical review of hand manifestations of cervical myelopathy, cervical radiculopathy, radial, ulnar, and median nerve neuropathies.

Cervical spondylotic myelopathy (CSM) is defined as compression of the spinal cord in the neck, resulting in problems with fine motor skills, hand numbness, pain or stiffness of the neck, and difficulty walking due to loss of balance. Brachial plexus (BP) neuropathies arise due to compression to any distal branches arising from C5-T1, whereas cervical radiculopathy involves compression at the nerve root in the neck. Such conditions can present with variable degrees of musculoskeletal pain, weakness, sensory changes, and reflex changes. The pronounced convergence in symptomatic manifestation within these conditions can pose a formidable challenge to clinicians, particularly in primary care. Thus, the primary objective of this paper is to enhance clarity and distinction among these pathological conditions. This objective is pursued through comprehensive delineation of the dermatomal and myotomal distributions characteristic of each condition. Furthermore, a meticulous examination is undertaken to elucidate physical indicators and maneuvers that exhibit a notably high sensitivity in detecting these conditions. Accurate diagnosis and treatment of each nerve pathology is important as long-term spinal cord compression and its roots may result in permanent disability and severely impact one's quality of life. As such, this systematic review serves as a guide that aids clinicians in differentiating the aforementioned conditions based on anatomy, physical exam findings, and imaging studies. Furthermore, this study aims to outline common peripheral nerve neuropathies in the upper extremities and ways to mitigate these pathologies using the least to most invasive treatment modalities.

Predicting colorectal cancer risk: a novel approach using anemia and blood test markers.

Colorectal cancer remains an important public health problem in the context of the COVID-19 (Corona virus disease 2019) pandemic. The decline in detection rates and delayed diagnosis of the disease necessitate the exploration of novel approaches to identify individuals with a heightened risk of developing colorectal cancer. The study aids clinicians in the rational allocation and utilization of healthcare resources, thereby benefiting patients, physicians, and the healthcare system. The present study retrospectively analyzed the clinical data of colorectal cancer cases diagnosed at the Affiliated Hospital of Guilin Medical University from September 2022 to September 2023, along with a control group. The study employed univariate and multivariate logistic regression as well as LASSO (Least absolute shrinkage and selection operator) regression to screen for predictors of colorectal cancer risk. The optimal predictors were selected based on the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. These predictors were then utilized in constructing a Nomogram Model for predicting colorectal cancer risk. The accuracy of the risk prediction Nomogram Model was assessed through calibration curves, ROC curves, and decision curve analysis (DCA) curves. Clinical data of 719 patients (302 in the case group and 417 in the control group) were included in this study. Based on univariate logistic regression analysis, there is a correlation between Body Mass Index (BMI), red blood cell count (RBC), anemia, Mean Corpuscular Volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), platelet count (PLT), Red Cell Distribution Width-Standard Deviation (RDW-SD), and the incidence of colorectal cancer. Based on the findings of multivariate logistic regression analysis, the variables of BMI and RBC exhibit a decrease, while anemia and PLT demonstrate an increase, all of which are identified as risk factors for the occurrence of colorectal cancer. LASSO regression selected BMI, RBC, anemia, and PLT as prediction factors. LASSO regression and multivariate logistic regression analysis yielded the same results. A nomogram was constructed based on the 4 prediction factors identified by LASSO regression analysis to predict the risk of colorectal cancer. The AUC of the nomogram was 0.751 (95% CI, OR: 0.708-0.793). The calibration curves in the validation and training sets showed good performance, indicating that the constructed nomogram model has good predictive ability. Additionally, the DCA demonstrated that the nomogram model has diagnostic accuracy. The Nomogram Model offers precise prognostications regarding the likelihood of Colorectal Cancer in patients, thereby helping healthcare professionals in their decision-making processes and promoting the rational categorization of patients as well as the allocation of medical resources.