Machine learning for predicting opioid use disorder from healthcare data: A systematic review.

INTRODUCTION: The US opioid epidemic has been one of the leading causes of injury-related deaths according to the CDC Injury Center. The increasing availability of data and tools for machine learning (ML) resulted in more researchers creating datasets and models to help analyze and mitigate the crisis. This review investigates peer-reviewed journal papers that applied ML models to predict opioid use disorder (OUD). The review is split into two parts. The first part summarizes the current research in OUD prediction with ML. The second part evaluates how ML techniques and processes were used to achieve these results and suggests improvements to refine further attempts to use ML for OUD prediction. METHODS: The review includes peer-reviewed journal papers published on or after 2012 that use healthcare data to predict OUD. We searched Google Scholar, Semantic Scholar, PubMed, IEEE Xplore, and Science.gov in September of 2022. Data extracted includes the study's goal, dataset used, cohort selected, types of ML models created, model evaluation metrics, and the details of the ML tools and techniques used to create the models. RESULTS: The review analyzed 16 papers. Three papers created their dataset, five used a publicly available dataset, and the remaining eight used a private dataset. Cohort size ranged from the low hundreds to over half a million. Six papers used one type of ML model, and the remaining ten used up to five different ML models. The reported ROC AUC was higher than 0.8 for all but one of the papers. Five papers used only non-interpretable models, and the other 11 used interpretable models exclusively or in combination with non-interpretable ones. The interpretable models were the highest or second-highest ROC AUC values. Most papers did not sufficiently describe the ML techniques and tools used to produce their results. Only three papers published their source code. CONCLUSIONS: We found that while there are indications that ML methods applied to OUD prediction may be valuable, the lack of details and transparency in creating the ML models limits their usefulness. We end the review with recommendations to improve studies on this critical healthcare subject.

Rapid deployment of inexpensive open-source orbital shakers in support of high-throughput screening.

Open-source projects continue to grow in popularity alongside open-source educational resources, software, and hardware tools. The impact of this increased availability of open-source technologies is that end users are empowered to have greater control over the tools that they work with. This trend extends in the life science laboratory space, where new open-source projects are routinely being published that allow users to build and modify scientific equipment specifically tailored to their needs, often at a reduced cost from equivalent commercial offerings. Recently, we identified a need for a compact orbital shaker that would be usable in temperature and humidity-controlled incubators to support the development and execution of a high-throughput suspension cell-based assay. Based on the requirements provided by staff biologists, an open-source project known as the DIYbio orbital shaker was identified on Thingiverse, then quickly prototyped and tested. The initial orbital shaker prototype based on the DIYbio design underwent an iterative prototyping and design process that proved to be straightforward due to the open-source nature of the project. The result of these efforts has been the successful initial deployment of ten shakers as of August 2021. This afforded us the scalability and efficacy needed to complete a large-scale screening campaign in less time and at less cost than if we purchased larger, less adaptable orbital shakers. Lessons learned from prototyping, modifying, validating, deploying and maintaining laboratory devices based on an open-source design in support of a full-scale drug discovery high-throughput screening effort are described within this manuscript.

Automating a Magnetic 3D Spheroid Model Technology for High-Throughput Screening.

Affordable and physiologically relevant three-dimensional (3D) cell-based assays used in high-throughput screening (HTS) are on the rise in early drug discovery. These technologies have been aided by the recent adaptation of novel microplate treatments and spheroid culturing techniques. One such technology involves the use of nanoparticle (NanoShuttle-PL) labeled cells and custom magnetic drives to assist in cell aggregation to ensure rapid 3D structure formation after the cells have been dispensed into microtiter plates. Transitioning this technology from a low-throughput manual benchtop application, as previously published by our lab, into a robotically enabled format achieves orders of magnitude greater throughput but required the development of specialized support hardware. This effort included in-house development, fabrication, and testing of ancillary devices that assist robotic handing and high-precision placement of microtiter plates into an incubator embedded with magnetic drives. Utilizing a "rapid prototyping" approach facilitated by cloud-based computer-aided design software, we built the necessary components using hobby-grade 3D printers with turnaround times that rival those of traditional manufacturing/development practices at a substantially reduced cost. This approach culminated in a first-in-class HTS-compatible 3D system in which we have coupled 3D bioprinting to a fully automated HTS robotic platform utilizing our novel magnetic incubator shelf assemblies.