Implementation of a Universal Screening Process for Substance Use in Pregnancy.

OBJECTIVE: To implement a standardized universal substance use screening process in an outpatient prenatal clinic at an urban tertiary care hospital. METHODS: Using a quality-improvement framework that involved process modeling, stakeholder analyses, and plan-do-study-act cycles, we implemented universal substance use screening for prenatal patients using a modified 5Ps screening tool (Parents, Peers, Partner, Past, Present). Implementation included an operational workflow based on the SBIRT (Screening, Brief Intervention, Referral to Treatment) model. The primary outcome measure was percentage of patients who were screened for substance use, with a goal of 90% screened. Secondary outcome measures were percentage who screened positive and percentage of the time a positive screen resulted in documentation of a brief intervention by a health care practitioner. RESULTS: Over a 19-month implementation period, 733 patient encounters were sampled. A substance use screen was completed in 618 (84%). We exceeded our goal of screening 90% of eligible patients for the final 6 months of data collection. Of the 618 completed screens, 124 (20%) screened positive. Health care practitioner documentation of brief interventions for patients with a positive screen reached 80% in the final phase of implementation, but then declined to 50% by the completion of the study period. CONCLUSION: A sustainable and generalizable process to carry out substance use screening within a large prenatal practice is feasible, and assisted with identification of patients not known to be at risk. Further efforts are needed to evaluate how to sustain health care practitioner documentation of intervention in response to positive screens.

Lesion Imaging and Target Detection in Multiple Scattering (LITMUS) Media.

We present an ultrasound algorithm [lesion imaging and target detection in multiple scattering (LITMUS)] suited to image lesions (hypoechoic) or targets (hyperechoic) in highly complex structures. In such media, standard ultrasound imaging techniques fail to detect lesions or targets due to aberrations and the loss of linearity between propagation distance and propagation time, caused by multiple scattering of ultrasound waves. The present algorithm (LITMUS) has the capability to predict the location as well as the size of such lesions/targets by using the multiple scattered ultrasound signals to its advantage. In this experimental and computational study, we use an ultrasound linear array. Lesions/targets are embedded at varying depths inside multiple scattering media with varying density of scatterers. In the simulations, plastic scatterers are used as the source of multiple scattering in a propagation medium (water). In the experiments, melamine sponges are used, with air alveoli as the scattering source. For multiple locations along the transducer, the incoherent backscattered intensity of the backscattered signals is extracted and the linear growth of the diffusive halo over time is tracked. Sudden changes in this growth indicate the presence of a region with reduced heterogeneity, indicative of the presence of a lesion/target. This methodology is combined with a depression detection algorithm to predict the size and location of the lesion/targets with high fidelity, despite the presence of strong heterogeneity and multiple scattering.

Artificial neural network to estimate micro-architectural properties of cortical bone using ultrasonic attenuation: A 2-D numerical study.

The goal of this study is to estimate micro-architectural parameters of cortical porosity such as pore diameter (φ), pore density (ρ) and porosity (ν) of cortical bone from ultrasound frequency dependent attenuation using an artificial neural network (ANN). First, heterogeneous structures with controlled pore diameters and pore densities (mono-disperse) were generated, to mimic simplified structure of cortical bone. Then, more realistic structures were obtained from high resolution CT scans of human cortical bone. 2-D finite-difference time-domain simulations were conducted to calculate the frequency-dependent attenuation in the 1-8 MHz range. An ANN was then trained with the ultrasonic attenuation at different frequencies as the input feature vectors while the output was set as the micro-architectural parameters (pore diameter, pore density and porosity). The ANN is composed of three fully connected dense layers with 24, 12 and 6 neurons, connected to the output layer. The dataset was trained over 6000 epochs with a batch size of 16. The trained ANN exhibits the ability to predict the micro-architectural parameters with high accuracy and low losses. ANN approaches could potentially be used as a tool to help inform physics-based modelling of ultrasound propagation in complex media such as cortical bone. This will lead to the solution of inverse-problems to retrieve bone micro-architectural parameters from ultrasound measurements for the non-invasive diagnosis and monitoring osteoporosis.

Predicting Structural Properties of Cortical Bone by Combining Ultrasonic Attenuation and an Artificial Neural Network (ANN): 2-D FDTD Study.

The goal of this paper is to predict the micro-architectural parameters of cortical bone such as pore diameter (ϕ) and porosity (v) from ultrasound attenuation measurements using an artificial neural network (ANN). Slices from a 3-D CT scan of human femur are obtained. The micro-architectural parameters of porosity such as average pore size and porosity are calculated using image processing. When ultrasound waves propagate in porous structures, attenuation is observed due to scattering. Two-dimensional finite-difference time-domain simulations are carried out to obtain frequency dependent attenuation in those 2D structures. An artificial neural network (ANN) is then trained with the input feature vector as the frequency dependent attenuation and output as pore diameter (ϕ) and porosity (v). The ANN is composed of one input layer, 3 hidden layers and one output layer, all of which are fully connected. 340 attenuation data sets were acquired and trained over 2000 epochs with a batch size of 32. Data was split into train, validation and test. It was observed that the ANN predicted the micro-architectural parameters of the cortical bone with high accuracies and low losses with a minimum R2 (goodness of fit) value of 0.95. ANN approaches could potentially help inform the solution of inverse-problems to retrieve bone porosity from ultrasound measurements. Ultimately, those inverse-problems could be used for the non-invasive diagnosis and monitoring of osteoporosis.