AIM: This systematic review aimed to investigate the performance of AI systems in identifying dental anomalies in paediatric dentistry and compare it with human performance.
DESIGN: A systematic search of Scopus, PubMed and Google Scholar was conducted from 2012 to 2022. Inclusion criteria were based on problem/patient/population, intervention/indicator, comparison and outcome scheme and specific keywords related to AI, DL, paediatric dentistry, dental anomalies, supernumerary and mesiodens. Six of 3918 initial pool articles were included, assessing nine DL sub-systems that used panoramic radiographs or cone-beam computed tomography. Article quality was assessed using QUADAS-2.
RESULTS: Artificial intelligence systems based on DL algorithms showed promising potential in enhancing the speed and accuracy of dental anomaly detection, with an average of 85.38% accuracy and 86.61% sensitivity. Human performance, however, outperformed AI systems, achieving 95% accuracy and 99% sensitivity. Limitations included a limited number of articles and data heterogeneity.
CONCLUSION: The potential of AI systems employing DL algorithms is highlighted in detecting dental anomalies in paediatric dentistry. Further research is needed to address limitations, explore additional anomalies and establish the broader applicability of AI in paediatric dentistry.
METHODS: Articles published between 2010 and 2023 were searched from five electronic databases. 59 papers were included for analysis with regards to: i). types of motion tested (functional vs. purposeful ankle movement); ii) types of biomechanical parameters measured (kinetic vs kinematic); iii) types of sensor systems used (lab-based vs field-based); and, iv) AI techniques used.
FINDINGS: Most studies (83.1%) examined biomechanics during functional motion. Single kinematic parameter, specifically ankle range of motion, could obtain accuracy up to 100% in identifying injury status. Wearable sensor exhibited high reliability for use in both laboratory and on-field/clinical settings. AI algorithms primarily utilized electromyography and joint angle information as input data. Support vector machine was the most used supervised learning algorithm (18.64%), while artificial neural network demonstrated the highest accuracy in eight studies.
INTERPRETATIONS: The potential for remote patient monitoring is evident with the adoption of field-based devices. Nevertheless, AI-based sensors are underutilized in detecting ankle motions at risk of sprain. We identify three key challenges: sensor designs, the controllability of AI models, and the integration of AI-sensor models, providing valuable insights for future research.
DESIGN: De-identified images were provided retrospectively or collected prospectively by IVF clinics using the artificial intelligence model in clinical practice. A total of 9359 images were provided by 18 IVF clinics across six countries, from 4709 women who underwent IVF between 2011 and 2021. Main outcome measures included clinical pregnancy outcome (fetal heartbeat at first ultrasound scan), embryo morphology score, and/or pre-implantation genetic testing for aneuploidy (PGT-A) results.
RESULTS: A positive linear correlation of artificial intelligence scores with pregnancy outcomes was found, and up to a 12.2% reduction in time to pregnancy (TTP) was observed when comparing the artificial intelligence model with standard morphological grading methods using a novel simulated cohort ranking method. Artificial intelligence scores were significantly correlated with known morphological features of embryo quality based on the Gardner score, and with previously unknown morphological features associated with embryo ploidy status, including chromosomal abnormalities indicative of severity when considering embryos for transfer during IVF.
CONCLUSION: Improved methods for evaluating artificial intelligence for embryo selection were developed, and advantages of the artificial intelligence model over current grading approaches were highlighted, strongly supporting the use of the artificial intelligence model in a clinical setting.
METHODS: Data were collected and analysed over three months from January 2023 using an ontology-based information extraction system (Semantic Hub). The system identified patient "stories" and extracted themes from online posts from January 2013 to March 2023, focusing on Korea and Taiwan by filtering the geographic location of users, the language used, and the local online platforms. Extracted texts were structured into knowledge graphs and analysed descriptively.
RESULTS: The patient voice was identified in 133,857 messages (9,620 patients) from the Naver online platform in Korea and included internet chat forums focused on macular degeneration. The most important factors for AMD treatments were effectiveness (1,632/3,401 mentions; 48%), price and access to insurance (33%), tolerability (10%) and doctor and clinic recommendations (9%). Treatment burden associated with intravitreal injection of vascular endothelial growth factor inhibitors related to tolerability (254/942 mentions; 27%), financial burden (20%), hospital selection (18%) and emotional burden (14%). In Taiwan, 444 messages were identified from Facebook, YouTube and Instagram. The success of treatment was judged by improvements in visual acuity (20/121 mentions; 16.5%), effect on oedema (10.7%), less distortion (9.1%) and inhibition of angiogenesis (5.8%). Tolerability concerns were rarely mentioned (26/440 mentions; 5.9%).
CONCLUSIONS: Digital Listening using Semantic-NLP can provide real-world insights from large amounts of internet data quickly and with low human labour cost. This allows healthcare companies to respond to the unmet needs of patients for effective and safe treatment and improved patient quality of life throughout the product lifecycle.
RESULTS: The oocyte AI model demonstrated area under the curve (AUC) up to 0.65 on two blind test datasets. High sensitivity for predicting competent oocytes (83-88%) was offset by lower specificity (26-36%). Exclusion of confounding biological variables (male factor infertility and maternal age ≥35 years) improved AUC up to 14%, primarily due to increased specificity. AI score correlated with size of the zona pellucida and perivitelline space, and ooplasm appearance. AI score also correlated with blastocyst expansion grade and morphological quality. The sum of AI scores from oocytes in group culture images predicted the formation of two or more usable blastocysts (AUC 0.77).
CONCLUSION: An AI model to evaluate oocyte competence was developed using federated learning, representing an essential step in protecting patient data. The AI model was significantly predictive of oocyte competence, as defined by usable blastocyst formation, which is a critical factor for IVF success. Potential clinical utility ranges from selective oocyte fertilization to guiding treatment decisions regarding additional rounds of oocyte retrieval.
DESIGN: In total, 10,677 oocyte images with associated metadata were collected prospectively by eight IVF clinics across six countries. AI training used federated learning, where data were retained on regional servers to comply with data privacy laws. The final AI model required a single image as input to evaluate oocyte competence, which was defined by the formation of a usable blastocyst (≥expansion grade 3 by day 5 or 6 post ICSI).