Deep learning predicts the differentiation of kidney organoids derived from human induced pluripotent stem cells

Kidney organoids derived from human pluripotent stem cells (hPSCs) contain multilineage nephrogenic progenitor cells and can recapitulate the development of the kidney. Kidney organoids derived from hPSCs have the potential to be applied in regenerative medicine as well as renal disease modeling, drug screening, and nephrotoxicity testing. Despite biotechnological advances, individual differences in morphological and growth characteristics among kidney organoids need to be addressed before clinical and commercial application. In this study, we hypothesized that an automated noninvasive method based on deep learning of bright-field images of kidney organoids can predict their differentiation status. Methods: Bright-field images of kidney organoids were collected on day 18 after differentiation. To train convolutional neural networks (CNNs), we utilized a transfer learning approach. CNNs were trained to predict the differentiation of kidney organoids on bright-field images based on the messenger RNA expression of renal tubular epithelial cells as well as podocytes. Results: The best prediction model was DenseNet121 with a total Pearson correlation coefficient score of 0.783 on a test dataset. W classified the kidney organoids into two categories: organoids with above-average gene expression (Positive) and those with below-average gene expression (Negative). Comparing the best-performing CNN with human-based classifiers, the CNN algorithm had a receiver operating characteristic-area under the curve (AUC) score of 0.85, while the experts had an AUC score of 0.48. Conclusion: These results confirmed our original hypothesis and demonstrated that our artificial intelligence algorithm can successfully recognize the differentiation status of kidney organoids.

Recent Advances in the Application of Artificial Intelligence in Otorhinolaryngology-Head and Neck Surgery

This study presents an up-to-date survey of the use of artificial intelligence (AI) in the field of otorhinolaryngology, considering opportunities, research challenges, and research directions. We searched PubMed, the Cochrane Central Register of Controlled Trials, Embase, and the Web of Science. We initially retrieved 458 articles. The exclusion of non-English publications and duplicates yielded a total of 90 remaining studies. These 90 studies were divided into those analyzing medical images, voice, medical devices, and clinical diagnoses and treatments. Most studies (42.2%, 38/90) used AI for image-based analysis, followed by clinical diagnoses and treatments (24 studies). Each of the remaining two subcategories included 14 studies. Machine learning and deep learning have been extensively applied in the field of otorhinolaryngology. However, the performance of AI models varies and research challenges remain.

The Assessment and Consideration about Feasibility of eFAST Exam in Medical School Students in Korea

PURPOSE: Emergency physicians usually perform ultrasound exams for patients with multiple injuries. Extended focused assessment with sonography for trauma (eFAST) can reveal injuries from chest to the abdomen immediately. However, in Korea, the curriculums of medical schools do not currently include eFAST. We have devised a study to assess the feasibility of the eFAST exam in medical school students. METHODS: This study was conducted in students in their fifth year out of six years of medical school, over 11 weeks. Four Emergency Medicine specialists trained the students over 4 hours, tested the students, and conducted a questionnaire. RESULTS: Average age of students was 25.9+/-2.6 years, and 24 were male and 20 were female. Mean success rate of 17 components on the eFAST exam was 95.9% (94.6%-97.4%). The success rate of transverse view of aorta, transverse view of bladder, lung sliding sign, and sea-shore sign of both anterior chest walls was 100%, scanning the spleen and attaching the probe to a body surface were 75%, 86.3%. Total time consumption was 449.0+/-22.2 seconds. The questionnaires showed that the eFAST exam for the splenorenal recess, spleen, and left upper quadrant lung was difficult. CONCLUSION: We found that medical students in Korea could perform the eFAST exam by themselves after 4 hours education. In the future, these findings can be helpful in development of an eFAST education program for medical students.

Guidewire Entrapment During Central Venous Catheterization

Central venous catheterization is common in the emergency department for monitoring of CVP (central venous pressure), fluid administration, and drug infusions. However, the insertion of a central venous catheter is a technically challenging procedure with known risks and complications. A 94-year-old woman was transferred to an emergency department due to difficulties in removing the guidewire during central catheter insertion through the right subclavian vein. A focused bedside ultrasound showed that the guidewire was improperly positioned in the right internal jugular vein. Upon computed tomographic evaluation, the guidewire perforated the right subclavian vein, looped in the mediastinum, reentered the right internal jugular vein toward the right jugular foramen, and was removed by surgery. In conclusion, as catheter and guidewire entrapment are well-known potential complications of central venous catheterization, when resistance is encountered at any stage of central venous catheterization (especially when removing the entrapped catheter or guidewire) the procedure should be stopped and evaluated with imaging assistance. Clinicians should be aware of more complicated sequelae caused by blunt removal of an entrapped catheter and guidewire, despite its low probability.