Search

Cities undergo constant change and their industrial spaces are particularly impacted on by new technologies, changing industries and the vagaries of the global and national economies. This paper reports on an investigation of the changing commercial and industrial landscape in Cape Town, South Africa, by focussing on commercial park spaces, specifically those that have been purposely named as business parks, industrial parks and office parks. The study is grounded in the literature on the changing spatial economy of Cape Town. A comprehensive database of commercial parks was created from a host of online sources and documents as well as a field survey. The spatial distribution of commercial parks was analysed in relation to the 11 districts mapped by a previous spatial economic study of the city. It was found that while the commercial parks were predominantly located in four districts, a nuanced locational analysis of the three different types of parks provided evidence of subtle differences in their distribution. The commercial parks are located both in traditional industrial suburbs and in newer commercial spaces and, although most are new developments, some are housed in old industrial spaces that have been converted. Almost half of the commercial parks are managed as sectional title schemes which mirror the management model of gated communities. Purposely named commercial parks are essentially a post-apartheid phenomenon as the majority are post-1994 developments.

Analysis of neighborhood environments is important for understanding inequality. Few studies, however, use direct measures of the visible characteristics of neighborhood conditions, despite their theorized importance in shaping individual and community well-being, because collecting data on the physical conditions of places across neighborhoods and cities and over time has required extensive time and labor. The authors introduce systematic social observation at scale (SSO@S), a pipeline for using visual data, crowdsourcing, and computer vision to identify visible characteristics of neighborhoods at a large scale. The authors implement SSO@S on millions of street-level images across three physically distinct cities—Boston, Detroit, and Los Angeles—from 2007 to 2020 to identify trash across space and over time. The authors evaluate the extent to which this approach can be used to assist with systematic coding of street-level imagery through cross-validation and out-of-sample validation, class-activation mapping, and comparisons with other sources of observed neighborhood characteristics. The SSO@S approach produces estimates with high reliability that correlate with some expected demographic characteristics but not others, depending on the city. The authors conclude with an assessment of this approach for measuring visible characteristics of neighborhoods and the implications for methods and research.

A decade of unprecedented progress in artificial intelligence (AI) has demonstrated the potential for many fields—including medicine—to benefit from the insights that AI techniques can extract from data. Here we survey recent progress in the development of modern computer vision techniques—powered by deep learning—for medical applications, focusing on medical imaging, medical video, and clinical deployment. We start by briefly summarizing a decade of progress in convolutional neural networks, including the vision tasks they enable, in the context of healthcare. Next, we discuss several example medical imaging applications that stand to benefit—including cardiology, pathology, dermatology, ophthalmology–and propose new avenues for continued work. We then expand into general medical video, highlighting ways in which clinical workflows can integrate computer vision to enhance care. Finally, we discuss the challenges and hurdles required for real-world clinical deployment of these technologies.

For newly diagnosed breast cancer, estrogen receptor status (ERS) is a key molecular marker used for prognosis and treatment decisions. During clinical management, ERS is determined by pathologists from immunohistochemistry (IHC) staining of biopsied tissue for the targeted receptor, which highlights the presence of cellular surface antigens. This is an expensive, time-consuming process which introduces discordance in results due to variability in IHC preparation and pathologist subjectivity. In contrast, hematoxylin and eosin (H&E) staining—which highlights cellular morphology—is quick, less expensive, and less variable in preparation. Here we show that machine learning can determine molecular marker status, as assessed by hormone receptors, directly from cellular morphology. We develop a multiple instance learning-based deep neural network that determines ERS from H&E-stained whole slide images (WSI). Our algorithm—trained strictly with WSI-level annotations—is accurate on a varied, multi-country dataset of 3,474 patients, achieving an area under the curve (AUC) of 0.92 for sensitivity and specificity. Our approach has the potential to augment clinicians’ capabilities in cancer prognosis and theragnosis by harnessing biological signals imperceptible to the human eye., Determination of estrogen receptor status (ERS) in breast cancer tissue requires immunohistochemistry, which is sensitive to the vagaries of sample processing and the subjectivity of pathologists. Here the authors present a deep learning model that determines ERS from H&E stained tissue, which could improve oncology decisions in under-resourced settings.

Bypassing nature’s evolutionary trajectory,de novoprotein generation—defined as creating artificial protein sequences from scratch—could enable breakthrough solutions for biomedical and environmental challenges. Viewing amino acid sequences as a language, we demonstrate that a deep learning-based language model can generate functional artificial protein sequences across families, akin to generating grammatically and semantically correct natural language sentences on diverse topics. Our protein language model is trained by simply learning to predict the next amino acid for over 280 million protein sequences from thousands of protein families, without biophysical or coevolutionary modeling. We experimentally evaluate model-generated artificial proteins on five distinct antibacterial lysozyme families. Artificial proteins show similar activities and catalytic efficiencies as representative natural lysozymes, including hen egg white lysozyme, while reaching as low as 44% identity to any known naturally-evolved protein. The X-ray crystal structure of an enzymatically active artificial protein recapitulates the conserved fold and positioning of active site residues found in natural proteins. We demonstrate our language model’s ability to be adapted to different protein families by accurately predicting the functionality of artificial chorismate mutase and malate dehydrogenase proteins. These results indicate that neural language models successfully performde novoprotein generation across protein families and may prove to be a tool to shortcut evolution.

Recent advances in self-supervised learning (SSL) have largely closed the gap with supervised ImageNet pretraining. Despite their success these methods have been primarily applied to unlabeled ImageNet images, and show marginal gains when trained on larger sets of uncurated images. We hypothesize that current SSL methods perform best on iconic images, and struggle on complex scene images with many objects. Analyzing contrastive SSL methods shows that they have poor visual grounding and receive poor supervisory signal when trained on scene images. We propose Contrastive Attention-Supervised Tuning (CAST) to overcome these limitations. CAST uses unsupervised saliency maps to intelligently sample crops, and to provide grounding supervision via a Grad-CAM attention loss. Experiments on COCO show that CAST significantly improves the features learned by SSL methods on scene images, and further experiments show that CAST-trained models are more robust to changes in backgrounds. Our code is available at https://github.com/salesforce/CAST/.

Generative modeling for protein engineering is key to solving fundamental problems in synthetic biology, medicine, and material science. We pose protein engineering as an unsupervised sequence generation problem in order to leverage the exponentially growing set of proteins that lack costly, structural annotations. We train a 1.2B-parameter language model, ProGen, on ∼280M protein sequences conditioned on taxonomic and keyword tags such as molecular function and cellular component. This provides ProGen with an unprecedented range of evolutionary sequence diversity and allows it to generate with fine-grained control as demonstrated by metrics based on primary sequence similarity, secondary structure accuracy, and conformational energy.