Automated prediction of extubation success in extremely preterm infants: the APEX multicenter study.

BACKGROUND: Extremely preterm infants are frequently subjected to mechanical ventilation. Current prediction tools of extubation success lacks accuracy. METHODS: Multicenter study including infants with birth weight ≤1250 g undergoing their first extubation attempt. Clinical data and cardiorespiratory signals were acquired before extubation. Primary outcome was prediction of extubation success. Automated analysis of cardiorespiratory signals, development of clinical and cardiorespiratory features, and a 2-stage Clinical Decision-Balanced Random Forest classifier were used. A leave-one-out cross-validation was done. Performance was analyzed by ROC curves and determined by balanced accuracy. An exploratory analysis was performed for extubations before 7 days of age. RESULTS: A total of 241 infants were included and 44 failed (18%) extubation. The classifier had a balanced accuracy of 73% (sensitivity 70% [95% CI: 63%, 76%], specificity 75% [95% CI: 62%, 88%]). As an additional clinical-decision tool, the classifier would have led to an increase in extubation success from 82% to 93% but misclassified 60 infants who would have been successfully extubated. In infants extubated before 7 days of age, the classifier identified 16/18 failures (specificity 89%) and 73/105 infants with success (sensitivity 70%). CONCLUSIONS: Machine learning algorithms may improve a balanced prediction of extubation outcomes, but further refinement and validation is required. IMPACT: A machine learning-derived predictive model combining clinical data with automated analyses of individual cardiorespiratory signals may improve the prediction of successful extubation and identify infants at higher risk of failure with a good balanced accuracy. Such multidisciplinary approach including medicine, biomedical engineering and computer science is a step forward as current tools investigated to predict extubation outcomes lack sufficient balanced accuracy to justify their use in future trials or clinical practice. Thus, this individualized assessment can optimize patient selection for future trials of extubation readiness by decreasing exposure of low-risk infants to interventions and maximize the benefits of those at high risk.

Using a Mobile App-Based Video Recommender System of Patient Narratives to Prepare Women for Breast Cancer Surgery: Development and Usability Study Informed by Qualitative Data.

BACKGROUND: Women diagnosed with breast cancer are often bombarded with information. Such information overload can lead to misunderstandings and hamper women's capacity for making informed decisions about their care. For women with breast cancer, this uncertainty is particularly severe in the period before surgery. Personalized narratives about others' experiences can help patients better understand the disease course, the quality and type of care to be expected, the clinical decision-making processes, and the strategies for coping. Existing resources and eHealth apps rarely include experiential information, and no tools exist that tailor information for individual preferences and needs-offering the right information at the right time and in the right format. Combining high-quality experiential evidence with novel technical approaches may contribute to patient-centered solutions in this area. OBJECTIVE: This study aims to design and seek preliminary feedback on a mobile app that will improve information access about surgery for patients with breast cancer, by drawing on a qualitative collection of personal narratives from a diverse sample of Canadian women and using video and audio recordings or audio recordings from the Canadian Health Experiences Research Network. METHODS: In a previous study, we conducted in-depth interviews with 35 Canadian women and used video and audio recordings or audio recordings to collect stories about the lived experiences of breast cancer. The participants highlighted the need for more specific information between diagnosis and surgery that was relevant to their personal situations and preferences. They also wanted to learn from other women's experiences. We worked with patients, clinicians, and informatics experts to develop a mobile app that provides access to tailored experiential information relevant to women's personal situations and preferences. We completed focus groups and qualitative interviews, conducted a further analysis of the original qualitative data, designed novel software using artificial intelligence, and sought preliminary feedback from users on a new app via focus groups and a survey. RESULTS: The secondary analysis of the breast cancer narratives revealed key themes and their interconnections relevant to the experience of surgery, including preparation, treatment decisions, aftercare, reconstruction, prostheses, lumpectomy and mastectomy, and complications. These themes informed the development of the structure and content of the app. We developed a recommender system within the app by using content matching (user and speaker profiles and user interests and video content) and collaborative filtering to identify clips marked as relevant by the user and by similar users. A 2-minute animated introductory video for users was developed. Pilot testing revealed generally positive responses regarding the content and value of this type of e-tool. CONCLUSIONS: Developing reliable, evidence-based tools and apps that are based on diverse collections of people's experiences of illness offers a novel approach to help manage the plethora of information that women face after a diagnosis of breast cancer.

Undersampling and Bagging of Decision Trees in the Analysis of Cardiorespiratory Behavior for the Prediction of Extubation Readiness in Extremely Preterm Infants.

Extremely preterm infants often require endotracheal intubation and mechanical ventilation during the first days of life. Due to the detrimental effects of prolonged invasive mechanical ventilation (IMV), clinicians aim to extubate infants as soon as they deem them ready.Unfortunately, existing strategies for prediction of extubation readiness vary across clinicians and institutions, and lead to high reintubation rates. We present an approach using Random Forest classifiers for the analysis of cardiorespiratory variability to predict extubation readiness. We address the issue of data imbalance by employing random undersampling of examples from the majority class before training each Decision Tree in a bag. By incorporating clinical domain knowledge, we further demonstrate that our classifier could have identified 71% of infants who failed extubation, while maintaining a success detection rate of 78%.

A semi-Markov chain approach to modeling respiratory patterns prior to extubation in preterm infants.

After birth, extremely preterm infants often require specialized respiratory management in the form of invasive mechanical ventilation (IMV). Protracted IMV is associated with detrimental outcomes and morbidities. Premature extubation, on the other hand, would necessitate reintubation which is risky, technically challenging and could further lead to lung injury or disease. We present an approach to modeling respiratory patterns of infants who succeeded extubation and those who required reintubation which relies on Markov models. We compare the use of traditional Markov chains to semi-Markov models which emphasize cross-pattern transitions and timing information, and to multi-chain Markov models which can concisely represent non-stationarity in respiratory behavior over time. The models we developed expose specific, unique similarities as well as vital differences between the two populations.