METHODS: This national-level, multicentre, prospective direct observational study was conducted in neonatal intensive care units (NICUs) of five public hospitals in Malaysia. Randomly selected nurses were directly observed during medication preparation and administration. Each observation was independently assessed for errors. Ten machine learning (ML) algorithms were applied with features derived from systematic reviews, incident reports, and expert consensus. Model performance, prioritising F1-score for MAEs, was evaluated using various measures. Feature importance was determined using the permutation-feature importance for robust comparison across ML algorithms.
RESULTS: A total of 1093 doses were administered to 170 neonates, with mean age and birth weight of 33.43 (SD ± 5.13) weeks and 1.94 (SD ± 0.95) kg, respectively. F1-scores for the ten models ranged from 76.15% to 83.28%. Adaptive boosting (AdaBoost) emerged as the best-performing model (F1-score: 83.28%, accuracy: 77.63%, area under the receiver operating characteristic: 82.95%, precision: 84.72%, sensitivity: 81.88% and negative predictive value: 64.00%). The most influential features in AdaBoost were the intravenous route of administration, working hours, and nursing experience.
CONCLUSIONS: This study developed and validated an ML-based model to predict the presence of MAEs among neonates in NICUs. AdaBoost was identified as the best-performing algorithm. Utilising the model's predictions, healthcare providers can potentially reduce MAE occurrence through timely interventions.
METHODS: This prospective comparative study was conducted from 2009 to 2012. Patients with ACL injuries who underwent knee arthroscopy and MRI were included in the study. Two radiologists who were blinded to the clinical history and arthroscopic findings reviewed the pre-arthroscopic MR images. The presence and type of meniscal tears on MRI and arthroscopy were recorded. Arthroscopic findings were used as the reference standard. The accuracy, sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) of MRI in the evaluation of meniscal tears were calculated.
RESULTS: A total of 65 patients (66 knees) were included. The sensitivity, specificity, accuracy, PPV, and NPV for the MRI diagnosis of lateral meniscal tears in our patients were 83, 97, 92, 96, and 90 %, respectively, whereas those for medial meniscus tears were 82, 92, 88, 82, and 88 %, respectively. There were five false-negative diagnoses of medial meniscus tears and four false-negative diagnoses of lateral meniscus tears. The majority of missed meniscus tears on MRI affected the peripheral posterior horns.
CONCLUSION: The sensitivity for diagnosing a meniscal tear was significantly higher when the tear involved more than one-third of the meniscus or the anterior horn. The sensitivity was significantly lower for tears located in the posterior horn and for vertically oriented tears. Therefore, special attention should be given to the peripheral posterior horns of the meniscus, which are common sites of injury that could be easily missed on MRI. The high NPVs obtained in this study suggest that MRI is a valuable tool prior to arthroscopy.