Postsurgical prescriptions for opioid naive patients and association with overdose and misuse: retrospective cohort study.
Brat, G. A; Agniel, D.; Beam, A.; Yorkgitis, B.; Bicket, M.; Homer, M.; Fox, K. P; Knecht, D. B; McMahill-Walraven, C. N; Palmer, N.; and others
Bmj, 360: j5790. 2018.
![Postsurgical prescriptions for opioid naive patients and association with overdose and misuse: retrospective cohort study [link]](//bibbase.org/img/filetypes/link.svg "paper")
paper
link
bibtex
abstract
@article{brat2018postsurgical,
title={Postsurgical prescriptions for opioid naive patients and association with overdose and misuse: retrospective cohort study},
author={Brat, Gabriel A and Agniel, Denis and Beam, Andrew and Yorkgitis, Brian and Bicket, Mark and Homer, Mark and Fox, Kathe P and Knecht, Daniel B and McMahill-Walraven, Cheryl N and Palmer, Nathan and others},
journal={Bmj},
volume={360},
pages={j5790},
abstract={ABSTRACT
OBJECTIVE
To quantify the effects of varying opioid prescribing patterns after surgery on dependence, overdose, or abuse in an opioid naive population.
DESIGN
Retrospective cohort study.
SETTING
Surgical claims from a linked medical and pharmacy administrative database of 37 651 619 commercially insured patients between 2008 and 2016.
PARTICIPANTS
1 015 116 opioid naive patients undergoing surgery.
MAIN OUTCOME MEASURES
Use of oral opioids after discharge as defined by
refills and total dosage and duration of use. The primary outcome was a composite of misuse identified by a diagnostic code for opioid dependence, abuse,
or overdose.
RESULTS
568 612 (56.0%) patients received postoperative opioids, and a code for abuse was identified for
5906 patients (0.6%, 183 per 100 000 person
years). Total duration of opioid use was the strongest predictor of misuse, with each refill and additional week of opioid use associated with an adjusted increase in the rate of misuse of 44.0% (95% confidence interval 40.8% to 47.2%, P<0.001), and 19.9% increase in hazard (18.5% to 21.4%, P<0.001), respectively.
CONCLUSIONS
Each refill and week of opioid prescription is associated with a large increase in opioid misuse among opioid naive patients. The data from this study suggest that duration of the prescription rather than dosage is more strongly associated with ultimate misuse in the early postsurgical period. The analysis quantifies the association of prescribing},
url_Paper={https://www.dropbox.com/s/1hneb1bii61xoql/brat_bmj_opioids_2018.pdf?dl=1},
year={2018},
keywords={Healthcare},
publisher={British Medical Journal Publishing Group}
}
ABSTRACT OBJECTIVE To quantify the effects of varying opioid prescribing patterns after surgery on dependence, overdose, or abuse in an opioid naive population. DESIGN Retrospective cohort study. SETTING Surgical claims from a linked medical and pharmacy administrative database of 37 651 619 commercially insured patients between 2008 and 2016. PARTICIPANTS 1 015 116 opioid naive patients undergoing surgery. MAIN OUTCOME MEASURES Use of oral opioids after discharge as defined by refills and total dosage and duration of use. The primary outcome was a composite of misuse identified by a diagnostic code for opioid dependence, abuse, or overdose. RESULTS 568 612 (56.0%) patients received postoperative opioids, and a code for abuse was identified for 5906 patients (0.6%, 183 per 100 000 person years). Total duration of opioid use was the strongest predictor of misuse, with each refill and additional week of opioid use associated with an adjusted increase in the rate of misuse of 44.0% (95% confidence interval 40.8% to 47.2%, P<0.001), and 19.9% increase in hazard (18.5% to 21.4%, P<0.001), respectively. CONCLUSIONS Each refill and week of opioid prescription is associated with a large increase in opioid misuse among opioid naive patients. The data from this study suggest that duration of the prescription rather than dosage is more strongly associated with ultimate misuse in the early postsurgical period. The analysis quantifies the association of prescribing
Big data and machine learning in health care.
Beam, A. L; and Kohane, I. S
Jama, 319(13): 1317–1318. 2018.
paper
link
bibtex
abstract
8 downloads
@article{beam2018big,
title={Big data and machine learning in health care},
author={Beam, Andrew L and Kohane, Isaac S},
journal={Jama},
volume={319},
number={13},
pages={1317--1318},
year={2018},
abstract={Nearly all aspectsof modern life are in some way being changed by big data and machine learning. Netflix knows what movies people like to watch and Google knows what people want to know based on their search histories. Indeed, Google has recently begun to replace much ofitsexistingnon–machinelearningtechnologywithmachine learning algorithms, and there is great optimism that these techniques can provide similar improvements across many sectors. },
url_Paper = {https://www.dropbox.com/s/8owxjsiked7f8hs/beam_kohane_jama_2018.pdf?dl=1},
keywords={Healthcare, Viewpoints},
publisher={American Medical Association}
}
Nearly all aspectsof modern life are in some way being changed by big data and machine learning. Netflix knows what movies people like to watch and Google knows what people want to know based on their search histories. Indeed, Google has recently begun to replace much ofitsexistingnon–machinelearningtechnologywithmachine learning algorithms, and there is great optimism that these techniques can provide similar improvements across many sectors.
Adversarial attacks against medical deep learning systems.
Finlayson, S. G; Chung, H. W.; Kohane, I. S; and Beam, A. L
arXiv preprint arXiv:1804.05296. 2018.
paper
link
bibtex
abstract
@article{finlayson2018adversarial,
title={Adversarial attacks against medical deep learning systems},
author={Finlayson, Samuel G and Chung, Hyung Won and Kohane, Isaac S and Beam, Andrew L},
journal={arXiv preprint arXiv:1804.05296},
url_Paper={https://www.dropbox.com/s/vt378etc6bpujuh/finlayson_adversarial_arxiv_2018.pdf?dl=1},
abstract={The discovery of adversarial examples has raised concerns about the practical deployment of deep learning systems. In this paper, we demonstrate that adversarial examples are capable of manip- ulating deep learning systems across three clinical domains. For each of our representative medical deep learning classifiers, both white and black box attacks were highly successful. Our models are representative of the current state of the art in medical computer vision and, in some cases, directly reflect architectures already see- ing deployment in real world clinical settings. In addition to the technical contribution of our paper, we synthesize a large body of knowledge about the healthcare system to argue that medicine may be uniquely susceptible to adversarial attacks, both in terms of monetary incentives and technical vulnerability. To this end, we outline the healthcare economy and the incentives it creates for fraud and provide concrete examples of how and why such attacks could be realistically carried out. We urge practitioners to be aware of current vulnerabilities when deploying deep learning systems in clinical settings, and encourage the machine learning community to further investigate the domain-specific characteristics of medical learning systems.},
keywords={Deep Learning, Adversarial Attacks, Healthcare},
year={2018}
}
The discovery of adversarial examples has raised concerns about the practical deployment of deep learning systems. In this paper, we demonstrate that adversarial examples are capable of manip- ulating deep learning systems across three clinical domains. For each of our representative medical deep learning classifiers, both white and black box attacks were highly successful. Our models are representative of the current state of the art in medical computer vision and, in some cases, directly reflect architectures already see- ing deployment in real world clinical settings. In addition to the technical contribution of our paper, we synthesize a large body of knowledge about the healthcare system to argue that medicine may be uniquely susceptible to adversarial attacks, both in terms of monetary incentives and technical vulnerability. To this end, we outline the healthcare economy and the incentives it creates for fraud and provide concrete examples of how and why such attacks could be realistically carried out. We urge practitioners to be aware of current vulnerabilities when deploying deep learning systems in clinical settings, and encourage the machine learning community to further investigate the domain-specific characteristics of medical learning systems.
Development of an algorithm to identify patients with physician-documented insomnia.
Kartoun, U.; Aggarwal, R.; Beam, A. L; Pai, J. K; Chatterjee, A. K; Fitzgerald, T. P; Kohane, I. S; and Shaw, S. Y
Scientific reports, 8(1): 7862. 2018.
paper
link
bibtex
abstract
2 downloads
@article{kartoun2018development,
title={Development of an algorithm to identify patients with physician-documented insomnia},
author={Kartoun, Uri and Aggarwal, Rahul and Beam, Andrew L and Pai, Jennifer K and Chatterjee, Arnaub K and Fitzgerald, Timothy P and Kohane, Isaac S and Shaw, Stanley Y},
journal={Scientific reports},
volume={8},
number={1},
pages={7862},
year={2018},
keywords={Healthcare},
url_Paper={https://www.dropbox.com/s/iwsykxorvq47kpb/kartoun_insomnia_scientificreports_2018.pdf?dl=1},
abstract={We developed an insomnia classification algorithm by interrogating an electronic medical records (EMR) database of 314,292 patients. The patients received care at Massachusetts General Hospital (MGH), Brigham and Women’s Hospital (BWH), or both, between 1992 and 2010. Our algorithm combined structured variables (such as International Classification of Diseases 9th Revision [ICD-9] codes, prescriptions, laboratory observations) and unstructured variables (such as text mentions of sleep and psychiatric disorders in clinical narrative notes). The highest classification performance of our algorithm was achieved when it included a combination of structured variables (billing codes for insomnia, common psychiatric conditions, and joint disorders) and unstructured variables (sleep disorders
and psychiatric disorders). Our algorithm had superior performance in identifying insomnia patients compared to billing codes alone (area under the receiver operating characteristic curve [AUROC] = 0.83 vs. 0.55 with 95% confidence intervals [CI] of 0.76–0.90 and 0.51–0.58, respectively). When applied to the 314,292-patient population, our algorithm classified 36,810 of the patients with insomnia, of which less than 17% had a billing code for insomnia. In conclusion, an insomnia classification algorithm that incorporates clinical notes is superior to one based solely on billing codes. Compared to traditional methods, our study demonstrates that a classification algorithm that incorporates physician notes can more accurately, comprehensively, and quickly identify large cohorts of insomnia patients.},
publisher={Nature Publishing Group}
}
We developed an insomnia classification algorithm by interrogating an electronic medical records (EMR) database of 314,292 patients. The patients received care at Massachusetts General Hospital (MGH), Brigham and Women’s Hospital (BWH), or both, between 1992 and 2010. Our algorithm combined structured variables (such as International Classification of Diseases 9th Revision [ICD-9] codes, prescriptions, laboratory observations) and unstructured variables (such as text mentions of sleep and psychiatric disorders in clinical narrative notes). The highest classification performance of our algorithm was achieved when it included a combination of structured variables (billing codes for insomnia, common psychiatric conditions, and joint disorders) and unstructured variables (sleep disorders and psychiatric disorders). Our algorithm had superior performance in identifying insomnia patients compared to billing codes alone (area under the receiver operating characteristic curve [AUROC] = 0.83 vs. 0.55 with 95% confidence intervals [CI] of 0.76–0.90 and 0.51–0.58, respectively). When applied to the 314,292-patient population, our algorithm classified 36,810 of the patients with insomnia, of which less than 17% had a billing code for insomnia. In conclusion, an insomnia classification algorithm that incorporates clinical notes is superior to one based solely on billing codes. Compared to traditional methods, our study demonstrates that a classification algorithm that incorporates physician notes can more accurately, comprehensively, and quickly identify large cohorts of insomnia patients.
Opportunities in machine learning for healthcare.
Ghassemi, M.; Naumann, T.; Schulam, P.; Beam, A. L; and Ranganath, R.
arXiv preprint arXiv:1806.00388. 2018.
paper
link
bibtex
abstract
2 downloads
@article{ghassemi2018opportunities,
title={Opportunities in machine learning for healthcare},
author={Ghassemi, Marzyeh and Naumann, Tristan and Schulam, Peter and Beam, Andrew L and Ranganath, Rajesh},
journal={arXiv preprint arXiv:1806.00388},
url_Paper={https://www.dropbox.com/s/gb43h8wf6rn1w4p/ghassemi_opportunities_arxiv_2018.pdf?dl=1},
abstract={Modern electronic health records (EHRs) provide data to answer clinically mean- ingful questions. The growing data in EHRs makes healthcare ripe for the use of machine learning. However, learning in a clinical setting presents unique chal- lenges that complicate the use of common machine learning methodologies. For example, diseases in EHRs are poorly labeled, conditions can encompass multiple underlying endotypes, and healthy individuals are underrepresented. This article serves as a primer to illuminate these challenges and highlights opportunities for members of the machine learning community to contribute to healthcare.
},
keywords={Healthcare, Reviews},
year={2018}
}
Modern electronic health records (EHRs) provide data to answer clinically mean- ingful questions. The growing data in EHRs makes healthcare ripe for the use of machine learning. However, learning in a clinical setting presents unique chal- lenges that complicate the use of common machine learning methodologies. For example, diseases in EHRs are poorly labeled, conditions can encompass multiple underlying endotypes, and healthy individuals are underrepresented. This article serves as a primer to illuminate these challenges and highlights opportunities for members of the machine learning community to contribute to healthcare.
Medical journals should embrace preprints to address the reproducibility crisis.
Oakden-Rayner, L.; Beam, A. L; and Palmer, L. J
2018.
paper
link
bibtex
abstract
@misc{oakden2018medical,
title={Medical journals should embrace preprints to address the reproducibility crisis},
author={Oakden-Rayner, Luke and Beam, Andrew L and Palmer, Lyle J},
year={2018},
keywords={Viewpoints, Healthcare},
url_Paper={https://www.dropbox.com/s/ndr1fok3e22mbvd/oakden-raynor_preprints_ije_2018.pdf?dl=0},
abstract={The world moves at an ever-increasing pace, and this is especially true for biomedical research. The expansion of research capacity enabled by high-performance computing and ubiquitous, high-speed internet have created a fertile environment for the rapid dissemination and iteration of new ideas. There has been an explosion of output in many biomedical fields, including -omics of many stripes, ‘in sil- ico’ research, very large cohort studies (‘biobanking’) and medical artificial intelligence.},
publisher={Oxford University Press}
}
The world moves at an ever-increasing pace, and this is especially true for biomedical research. The expansion of research capacity enabled by high-performance computing and ubiquitous, high-speed internet have created a fertile environment for the rapid dissemination and iteration of new ideas. There has been an explosion of output in many biomedical fields, including -omics of many stripes, ‘in sil- ico’ research, very large cohort studies (‘biobanking’) and medical artificial intelligence.
Artificial intelligence in healthcare.
Yu, K.; Beam, A. L; and Kohane, I. S
Nature biomedical engineering, 2(10): 719. 2018.
paper
link
bibtex
abstract
4 downloads
@article{yu2018artificial,
title={Artificial intelligence in healthcare},
author={Yu, Kun-Hsing and Beam, Andrew L and Kohane, Isaac S},
journal={Nature biomedical engineering},
volume={2},
number={10},
pages={719},
year={2018},
keywords={Deep Learning, Reviews, Healthcare},
url_Paper={https://www.dropbox.com/s/tji0n19y5rciza6/yu_ai_healthcare_naturebme_2019.pdf?dl=0},
abstract={Artificial intelligence (AI) is gradually changing medical practice. With recent progress in digitized data acquisition, machine learning and computing infrastructure, AI applications are expanding into areas that were previously thought to be only the province of human experts. In this Review Article, we outline recent breakthroughs in AI technologies and their biomedical applications, identify the challenges for further progress in medical AI systems, and summarize the economic, legal and social implications of AI in healthcare.},
publisher={Nature Publishing Group}
}
Artificial intelligence (AI) is gradually changing medical practice. With recent progress in digitized data acquisition, machine learning and computing infrastructure, AI applications are expanding into areas that were previously thought to be only the province of human experts. In this Review Article, we outline recent breakthroughs in AI technologies and their biomedical applications, identify the challenges for further progress in medical AI systems, and summarize the economic, legal and social implications of AI in healthcare.
Machine Learning for Health (ML4H) Workshop at NeurIPS 2018.
Antropova, N.; Beam, A.; Beaulieu-Jones, B. K; Chen, I.; Chivers, C.; Dalca, A.; Finlayson, S.; Fiterau, M.; Fries, J. A.; Ghassemi, M.; and others
arXiv preprint arXiv:1811.07216. 2018.
link
bibtex
@article{antropova2018machine,
title={Machine Learning for Health (ML4H) Workshop at NeurIPS 2018},
author={Antropova, Natalia and Beam, Andrew and Beaulieu-Jones, Brett K and Chen, Irene and Chivers, Corey and Dalca, Adrian and Finlayson, Sam and Fiterau, Madalina and Fries, Jason Alan and Ghassemi, Marzyeh and others},
journal={arXiv preprint arXiv:1811.07216},
keywords={Reviews},
year={2018}
}
![Challenges to the Reproducibility of Machine Learning Models in Health Care [link]](http://bibbase.org/img/filetypes/link.svg "paper")