Perspectives Regarding Medications for Opioid Use Disorder Among Individuals with Mental Illness.

Most people with co-occurring opioid use disorder (OUD) and mental illness do not receive effective medications for treating OUD. To investigate perspectives of adults in a publicly-funded mental health system regarding medications for OUD (MOUD), we conducted semi-structured telephone interviews with 13 adults with OUD (current or previous diagnosis) receiving mental health treatment. Themes that emerged included: perceiving or using MOUDs as a substitute for opioids or a temporary solution to prevent withdrawal symptoms; negative perceptions about methadone/methadone clinics; and viewing MOUD use as "cheating". Readiness to quit was important for patients to consider MOUDs. All participants were receptive to discussing MOUDs with their mental health providers and welcomed the convenience of receiving care for their mental health and OUD at the same location. In conclusion, clients at publicly-funded mental health clinics support MOUD treatment, signaling a need to expand access and build awareness of MOUDs in these settings.

Partial Convolution for Padding, Inpainting, and Image Synthesis.

Partial convolution weights convolutions with binary masks and renormalizes on valid pixels. It was originally proposed for image inpainting task because a corrupted image processed by a standard convolutional often leads to artifacts. Therefore, binary masks are constructed that define the valid and corrupted pixels, so that partial convolution results are only calculated based on valid pixels. It has been also used for conditional image synthesis task, so that when a scene is generated, convolution results of an instance depend only on the feature values that belong to the same instance. One of the unexplored applications for partial convolution is padding which is a critical component of modern convolutional networks. Common padding schemes make strong assumptions about how the padded data should be extrapolated. We show that these padding schemes impair model accuracy, whereas partial convolution based padding provides consistent improvements across a range of tasks. In this article, we review partial convolution applications under one framework. We conduct a comprehensive study of the partial convolution based padding on a variety of computer vision tasks, including image classification, 3D-convolution-based action recognition, and semantic segmentation. Our results suggest that partial convolution-based padding shows promising improvements over strong baselines.

Revisiting Image-Language Networks for Open-Ended Phrase Detection.

Most existing work that grounds natural language phrases in images starts with the assumption that the phrase in question is relevant to the image. In this paper we address a more realistic version of the natural language grounding task where we must both identify whether the phrase is relevant to an image and localize the phrase. This can also be viewed as a generalization of object detection to an open-ended vocabulary, introducing elements of few- and zero-shot detection. We propose an approach for this task that extends Faster R-CNN to relate image regions and phrases. By carefully initializing the classification layers of our network using canonical correlation analysis (CCA), we encourage a solution that is more discerning when reasoning between similar phrases, resulting in over double the performance compared to a naive adaptation on three popular phrase grounding datasets, Flickr30K Entities, ReferIt Game, and Visual Genome, with test-time phrase vocabulary sizes of 5K, 32K, and 159K, respectively.

Unsupervised Disentanglement of Pose, Appearance and Background from Images and Videos.

Unsupervised landmark learning is the task of learning semantic keypoint-like representations without the use of expensive input keypoint annotations. A popular approach is to factorize an image into a pose and appearance data stream, then to reconstruct the image from the factorized components. The pose representation should capture a set of consistent and tightly localized landmarks in order to facilitate reconstruction of the input image. Ultimately, we wish for our learned landmarks to focus on the foreground object of interest. However, the reconstruction task of the entire image forces the model to allocate landmarks to model the background. Using a motion-based foreground assumption, this work explores the effects of factorizing the reconstruction task into separate foreground and background reconstructions in an unsupervised way, allowing the model to condition only the foreground reconstruction on the unsupervised landmarks. Our experiments demonstrate that the proposed factorization results in landmarks that are focused on the foreground object of interest when measured against ground-truth foreground masks. Furthermore, the rendered background quality is also improved as ill-suited landmarks are no longer forced to model this content. We demonstrate this improvement via improved image fidelity in a video-prediction task. Code is available at https://github.com/NVIDIA/UnsupervisedLandmarkLearning.

A Qualitative Study of Primary Care Physicians' Experiences With Telemedicine During COVID-19.

BACKGROUND: Primary care practices rapidly adopted telemedicine visits because of the COVID-19 pandemic, but information on physician perspectives about these visits is lacking. METHODS: Fifteen semistructured interviews with practicing primary care physicians and physicians-in-training from a Southern California academic health system and group-model health maintenance organization were conducted to assess physician perspectives regarding the benefits and challenges of telemedicine. RESULTS: Physicians indicated that telemedicine improved patient access to care by providing greater convenience, although some expressed concern that certain groups of vulnerable patients were unable to navigate or did not possess the technology required to participate in telemedicine visits. Physicians noted that telemedicine visits offered more time for patient counseling, opportunities for better medication reconciliations, and the ability to see and evaluate patient home environments and connect with patient families. Challenges existed when visits required a physical examination. Physicians were very concerned about the loss of personal connections and touch, which they believed diminished expected rituals that typically strengthen physician-patient relationships. Physicians also observed that careful consideration to physician workflows may be needed to avoid physician burnout. CONCLUSIONS: Physicians reported that telemedicine visits offer new opportunities to improve the quality of patient care but noted changes to their interactions with patients. Many of these changes are positive, but it remains to be seen whether others such as lack of physical examination and loss of physical presence and touch adversely influence provider-patient communication, patient willingness to disclose concerns that may affect their care, and, ultimately, patient health outcomes.

Learning Discriminative Collections of Part Detectors for Object Recognition.

We propose a method to learn a diverse collection of discriminative parts from object bounding box annotations. Part detectors can be trained and applied individually, which simplifies learning and extension to new features or categories. We apply the parts to object category detection, pooling part detections within bottom-up proposed regions and using a boosted classifier with proposed sigmoid weak learners for scoring. On PASCAL VOC2010, we evaluate the part detectors' ability to discriminate and localize annotated keypoints and their effectiveness in detecting object categories.