Search for: All records

Mobile sensing data processed using machine learning models can passively and remotely assess mental health symptoms from the context of patients’ lives. Prior work has trained models using data from single longitudinal studies, collected from demographically homogeneous populations, over short time periods, using a single data collection platform or mobile application. The generalizability of model performance across studies has not been assessed. This study presents a first analysis to understand if models trained using combined longitudinal study data to predict mental health symptoms generalize across current publicly available data. We combined data from the CrossCheck (individuals living with schizophrenia) and StudentLife (university students) studies. In addition to assessing generalizability, we explored if personalizing models to align mobile sensing data, and oversampling less-represented severe symptoms, improved model performance. Leave-one-subject-out cross-validation (LOSO-CV) results were reported. Two symptoms (sleep quality and stress) had similar question-response structures across studies and were used as outcomes to explore cross-dataset prediction. Models trained with combined data were more likely to be predictive (significant improvement over predicting training data mean) than models trained with single-study data. Expected model performance improved if the distance between training and validation feature distributions decreased using combined versus single-study data. Personalization aligned each LOSO-CV participant with training data, but only improved predicting CrossCheck stress. Oversampling significantly improved severe symptom classification sensitivity and positive predictive value, but decreased model specificity. Taken together, these results show that machine learning models trained on combined longitudinal study data may generalize across heterogeneous datasets. We encourage researchers to disseminate collected de-identified mobile sensing and mental health symptom data, and further standardize data types collected across studies to enable better assessment of model generalizability. 

Digital biomarkers of mental health, created using data extracted from everyday technologies including smartphones, wearable devices, social media and computer interactions, have the opportunity to revolutionise mental health diagnosis and treatment by providing near-continuous unobtrusive and remote measures of behaviours associated with mental health symptoms. Machine learning models process data traces from these technologies to identify digital biomarkers. In this editorial, we caution clinicians against using digital biomarkers in practice until models are assessed for equitable predictions (‘model equity’) across demographically diverse patients at scale, behaviours over time, and data types extracted from different devices and platforms. We posit that it will be difficult for any individual clinic or large-scale study to assess and ensure model equity and alternatively call for the creation of a repository of open de-identified data for digital biomarker development. 

Investigations of organic lithic micro-residues have, over the last decade, shifted from entirely morphological observations using visible-light microscopy to compositional ones using scanning electron microscopy and Fourier-transform infrared microspectroscopy, providing a seemingly objective chemical basis for residue identifications. Contamination, though, remains a problem that can affect these results. Modern contaminants, accumulated during the post-excavation lives of artifacts, are pervasive, subtle, and even “invisible” (unlisted ingredients in common lab products). Ancient contamination is a second issue. The aim of residue analysis is to recognize residues related to use, but other types of residues can also accumulate on artifacts. Caves are subject to various taphonomic forces and organic inputs, and use-related residues can degrade into secondary compounds. This organic “background noise” must be taken into consideration. Here we show that residue contamination is more pervasive than is often appreciated, as revealed by our studies of Middle Palaeolithic artifacts from two sites: Lusakert Cave 1 in Armenia and Crvena Stijena in Montenegro. First, we explain how artifacts from Lusakert Cave 1, despite being handled following specialized protocols, were tainted by a modern-day contaminant from an unanticipated source: a release agent used inside the zip-top bags that are ubiquitous in the field and lab. Second, we document that, when non-artifact “controls” are studied alongside artifacts from Crvena Stijena, comparisons reveal that organic residues are adhered to both, indicating that they are prevalent throughout the sediments and not necessarily related to use. We provide suggestions for reducing contamination and increasing the reliability of residue studies. Ultimately, we propose that archaeologists working in the field of residue studies must start with the null hypothesis that miniscule organic residues reflect contamination, either ancient or modern, and systematically proceed to rule out all possible contaminants before interpreting them as evidence of an artifact’s use in the distant past.