Goto

Collaborating Authors

Results


A Hierarchy of Limitations in Machine Learning

arXiv.org Machine Learning

There is little argument about whether or not machine learning models are useful for applying to social systems. But if we take seriously George Box's dictum, or indeed the even older one that "the map is not the territory' (Korzybski, 1933), then there has been comparatively less systematic attention paid within the field to how machine learning models are wrong (Selbst et al., 2019) and seeing possible harms in that light. By "wrong" I do not mean in terms of making misclassifications, or even fitting over the'wrong' class of functions, but more fundamental mathematical/statistical assumptions, philosophical (in the sense used by Abbott, 1988) commitments about how we represent the world, and sociological processes of how models interact with target phenomena. This paper takes a particular model of machine learning research or application: one that its creators and deployers think provides a reliable way of interacting with the social world (whether that is through understanding, or in making predictions) without any intent to cause harm (McQuillan, 2018) and, in fact, a desire to not cause harm and instead improve the world, 1 for example as most explicitly in the various "{Data [Science], Machine Learning, Artificial Intelligence} for [Social] Good" initiatives, and more widely in framings around "fairness" or "ethics." I focus on the almost entirely statistical modern version of machine learning, rather than eclipsed older visions (see section 3). While many of the limitations I discuss apply to the use of machine learning in any domain, I focus on applications to the social world in order to explore the domain where limitations are strongest and stickiest.


Industry News

#artificialintelligence

Find here a listing of the latest industry news in genomics, genetics, precision medicine, and beyond. Updates are provided on a monthly basis. Sign-Up for our newsletter and never miss out on the latest news and updates. As 2019 came to an end, Veritas Genetics struggled to get funding due to concerns it had previously taken money from China. It was forced to cease US operations and is in talks with potential buyers. The GenomeAsia 100K Project announced its pilot phase with hopes to tackle the underrepresentation of non-Europeans in human genetic studies and enable genetic discoveries across Asia. Veritas Genetics, the start-up that can sequence a human genome for less than $600, ceases US operations and is in talks with potential buyers Veritas Genetics ceases US operations but will continue Veritas Europe and Latin America. It had trouble raising funding due to previous China investments and is looking to be acquired. Illumina loses DNA sequencing patents The European Patent ...


Training Deep Learning models with small datasets

arXiv.org Machine Learning

Miguel Romero BSc 1, Yannet Interian PhD 1, Timothy Solberg PhD 2, and Gilmer Valdes PhD 2 1 Master of Science in Data Science, University of San Francisco, San Francisco, CA 2 Department of Radiation Oncology, University of California San Francisco, San Francisco, CA December 17, 2019 Abstract The growing use of Machine Learning has produced significant advances in many fields. For image-based tasks, however, the use of deep learning remains challenging in small datasets. In this article, we review, evaluate and compare current state of the art techniques in training neural networks to elucidate which techniques work best for small datasets. We further propose a path forward for the improvement of model accuracy in medical imaging applications. We observed best results from: one cycle training, discriminative learning rates with gradual freezing and parameter modification after transfer learning. We also established that when datasets are small, transfer learning plays an important role beyond parameter initialization by reusing previously learned features. Surprisingly we observed that there is little advantage in using pre-trained networks in images from another part of the body compared to Imagenet. On the contrary, if images from the same part of the body are available then transfer learning can produce a significant improvement in performance with as little as 50 images in the training data. 1 Introduction The use of machine learning in medical imaging, radiation theranostics and medical physics applications has created tremendous opportunity with research that encompasses: quality assurance [1, 2, 3, 4, 5, 6], outcome prediction [7, 8, 9, 10, 11, 12, 13], segmentation [14, 15, 16, 17] or dosimetric prediction Equal contribution authors. Partially supported by the wicklow AI and medical research initiative at the Data institute.


Artificial intelligence better than humans at spotting lung cancer

#artificialintelligence

Researchers have used a deep-learning algorithm to detect lung cancer accurately from computed tomography scans. The results of the study indicate that artificial intelligence can outperform human evaluation of these scans. The condition is the leading cause of cancer-related death in the U.S., and early detection is crucial for both stopping the spread of tumors and improving patient outcomes. As an alternative to chest X-rays, healthcare professionals have recently been using computed tomography (CT) scans to screen for lung cancer. In fact, some scientists argue that CT scans are superior to X-rays for lung cancer detection, and research has shown that low-dose CT (LDCT) in particular has reduced lung cancer deaths by 20%.


Knowledge-based Biomedical Data Science 2019

arXiv.org Artificial Intelligence

Knowledge-based biomedical data science (KBDS) involves the design and implementation of computer systems that act as if they knew about biomedicine. Such systems depend on formally represented knowledge in computer systems, often in the form of knowledge graphs. Here we survey the progress in the last year in systems that use formally represented knowledge to address data science problems in both clinical and biological domains, as well as on approaches for creating knowledge graphs. Major themes include the relationships between knowledge graphs and machine learning, the use of natural language processing, and the expansion of knowledge-based approaches to novel domains, such as Chinese Traditional Medicine and biodiversity.


AI Model Identifies Risk of Mortality from Chest X-Rays

#artificialintelligence

An artificial intelligence (AI) model could help identify patients at an increased risk of long-term mortality based on their chest X-rays, according to the findings of a study published in JAMA Network Open. The convolutional neural network, named CXR-risk, found that 53% of people it identified as very high-risk for future heart attack, lung cancer or death died over 12 years. In comparison, fewer than 4% of those whom CXR-risk identified as very low-risk died over 12 years. The model was developed and trained on data sets from two clinical trials -- Prostate, Lung, Colorectal and Ovarian Cancer Screening Trial (PLCO) and National Lung Cancer Screening Trial (NLST) -- and found that mortality rates for patients with very high-risk scores had an 18- (PLCO) and 15-fold (NLST) higher mortality rate compared to those in the very low-risk category. "This is a new way to extract prognostic information from everyday diagnostic tests," said Michael Lu, M.D., MPH, radiology department at Massachusetts General Hospital.


Deep Learning to Assess Long-term Mortality From Chest Radiographs

#artificialintelligence

Question Is a convolutional neural network able to extract prognostic information from chest radiographs? Findings In this prognostic study of data from 2 randomized clinical trials (Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial [n 10 464] and National Lung Screening Trial [n 5493]), a convolutional neural network identified persons at high risk of long-term mortality based on their chest radiographs, even with adjustment for the radiologists' diagnostic findings and standard risk factors. Meaning Individuals at high risk of mortality based on chest radiography may benefit from prevention, screening, and lifestyle interventions. Importance Chest radiography is the most common diagnostic imaging test in medicine and may also provide information about longevity and prognosis. Objective To develop and test a convolutional neural network (CNN) (named CXR-risk) to predict long-term mortality, including noncancer death, from chest radiographs. Design, Setting, and Participants In this prognostic study, CXR-risk CNN development (n 41 856) and testing (n 10 464) used data from the screening radiography arm of the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO) (n 52 320), a community cohort of asymptomatic nonsmokers and smokers (aged 55-74 years) enrolled at 10 US sites from November 8, 1993, through July 2, 2001. External testing used data from the screening radiography arm of the National Lung Screening Trial (NLST) (n 5493), a community cohort of heavy smokers (aged 55-74 years) enrolled at 21 US sites from August 2002, through April 2004. Data analysis was performed from January 1, 2018, to May 23, 2019. Exposure Deep learning CXR-risk score (very low, low, moderate, high, and very high) based on CNN analysis of the enrollment radiograph.


Artificial intelligence better than humans at spotting lung cancer

#artificialintelligence

The condition is the leading cause of cancer-related death in the U.S., and early detection is crucial for both stopping the spread of tumors and improving patient outcomes. As an alternative to chest X-rays, healthcare professionals have recently been using computed tomography (CT) scans to screen for lung cancer. In fact, some scientists argue that CT scans are superior to X-rays for lung cancer detection, and research has shown that low-dose CT (LDCT) in particular has reduced lung cancer deaths by 20%. These errors typically delay the diagnosis of lung cancer until the disease has reached an advanced stage when it becomes too difficult to treat. New research may safeguard against these errors.