Goto

Collaborating Authors

6 AI Healthcare Solutions for Remote Patient Monitoring

#artificialintelligence

It's no secret that big tech companies like Amazon (AMZN), Microsoft (MSFT), and Alphabet (GOOG), the parent company of Google, are investing in digital healthcare. The market opportunity is pretty enticing when you consider that the U.S. alone spent $3.65 trillion on healthcare just last year. Google made the latest headline-grabbing move when it announced that it would buy wearables maker Fitbit (FIT) in a deal valued at $2.1 billion. Analysts have noted that the acquisition is part of the company's overall strategy to build an ambient intelligent system where Google is omnipresent. Another motive behind the purchase – pending regulatory approvals – is that Fitbit gives Google access to a treasure trove of healthcare data that it can feed to its London-based AI lab DeepMind or its life sciences subsidiary Verily, which is already collaborating on at least one AI healthcare device for remote patient monitoring.


6 AI Healthcare Solutions for Remote Patient Monitoring

#artificialintelligence

It's no secret that big tech companies like Amazon (AMZN), Microsoft (MSFT), and Alphabet (GOOG), the parent company of Google, are investing in digital healthcare. The market opportunity is pretty enticing when you consider that the U.S. alone spent $3.65 trillion on healthcare just last year. Google made the latest headline-grabbing move when it announced that it would buy wearables-maker Fitbit (FIT) in a deal valued at $2.1 billion. Analysts have noted that the acquisition is part of the company's overall strategy to build an ambient intelligent system where Google is omnipresent. Another motive behind the purchase – pending regulatory approvals – is that Fitbit gives Google access to a treasure trove of healthcare data that it can feed to its London-based AI lab DeepMind or its life sciences subsidiary Verily, which is already collaborating on at least one AI healthcare device for remote patient monitoring.


5 Ways Machine learning is Redefining Healthcare

#artificialintelligence

Machine learning (ML) is an application of artificial intelligence (AI) wherein the system looks at observations or data, such as examples, direct experience, or instruction, figures out patterns in data and predicts events in the future based on the examples that we provide. Machine learning is seeing more and more use across industries for various reasons: vast amounts of data are being captured and made available digitally; processing of large amounts of data has become cost-effective due to the increased computing power now available at affordable prices; and various open source frameworks, toolkits and libraries are available that can be used to build and execute ML applications. Specifically in healthcare, ML has led to exciting new developments that could redefine cancer diagnosis and treatment in the years to come. ML can increase access to treatment in developing countries which don't have enough specialist doctors that can treat certain diseases, it can improve the sensitivity of detection, add more value in treatment decisions, and it can help personalize treatment so that each patient gets the treatment that's best for them. In many cases they can even add to workflow efficiency in hospitals.


Better, faster, and even cheap

Science

Cryo–electron microscopy (cryo-EM) enables access to structures of proteins that were previously intractable, including large protein complexes such as the ribosome ([ 1 ][1]), integral membrane proteins ([ 2 ][2], [ 3 ][3]), and highly heterogeneous or conformationally dynamic systems ([ 4 ][4]). Each sample is a vitrified layer of protein suspended over a support film on an EM grid. Despite recent advances in cryo-EM (the so-called “resolution revolution”) ([ 5 ][5], [ 6 ][6]), major barriers persist, including loss of the highest-resolution information through electron beam damage and blurring from sample movement (which is most pronounced initially when the sample is least damaged). Typically, tens of thousands of images must be averaged to compensate for signal loss. On page 223 of this issue, Naydenova et al. ([ 7 ][7]) describe a new specimen support film (see the figure) that not only improves both the quality of images and the efficiency of collection, but also does so at a relatively low price. Like much of society, the strengths and weaknesses of cryo-EM have been highlighted by the coronavirus disease 2019 (COVID-19) pandemic. Structural biology has contributed to our understanding of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), with more than 330 structures deposited into the Protein Data Bank since early March. At the time of writing, 264 of those structures were produced by x-ray crystallography and 72 by cryo-EM. The astonishing throughput of crystallography was demonstrated in recent work on the main SARS-CoV-2 protease, where ∼1500 datasets were collected in a single day ([ 8 ][8]). In contrast, cryo-EM datasets typically require on the order of 24 hours of data collection, even though the process is largely automated ([ 9 ][9]). High-end microscopes are not only under extraordinary demand, but are also associated with very high capital and annual maintenance costs ([ 10 ][10]). These differences have made x-ray crystallography better suited to the rapid turnaround required by the pharmaceutical drug discovery pipeline ([ 11 ][11]). These timelines also present a barrier for academic research. Increasing throughput will lower costs and enable more researchers to use cryo-EM to solve structures and accelerate scientific discovery. Naydenova et al. propose a new specimen support film, dubbed “HexAuFoil,” that provides many advantages over conventional support films. In cryo-EM, images are typically taken of proteins embedded in a layer of vitrified ice suspended over a support films containing holes ∼1 to 2 µm in diameter. By making these holes smaller (200 to 300 nm in diameter) and packing them more tightly together (a nontrivial feat), they substantially increased data throughput. They obtained 200 images for every microscope stage movement (versus a typical 10 to 30 images) and propose that as many as 800 images could be obtained with an appropriately configured microscope. A second approach to increasing throughput that Naydenova et al. addressed is to increase the data quality by minimizing the information loss from both radiation damage and sample movement. They present a fascinating and exhaustive analysis showing that the primary cause of specimen movement during imaging is a buckling of the vitrified ice layer. On the basis of these insights, they created a support film with an optimal substrate thickness relative to the hole diameter, which reduces the total movement of the sample to <1 Å during the course of an exposure. They could mathematically extrapolate the data to a three-dimensional map before the onset of radiation damage. Standard support films have recently been used to obtain the first truly atomic-level cryo-EM reconstructions ([ 12 ][12], [ 13 ][13]) of apoferritin, a very stable test specimen. The advances developed by Naydenova et al. should bring this goal closer for less well-behaved proteins. Although software is also now available to correct for the effects of beam-induced movement ([ 14 ][14]), this approach does not provide the concomitant benefit of faster data collection. High-resolution cryo-EM is still in a state of active method development, its true potential not yet realized. Thus, it is exciting that Naydenova et al. , with one accessible, inexpensive hardware development, will allow all practitioners to acquire better images much more rapidly as soon as the grids become commercially available. The promise is of a future of high-resolution structures of a wide range of proteins, in an ensemble of conformational or compositional states ([ 15 ][15]), produced with much higher throughput. 1. [↵][16]1. A. Brown, 2. S. Shao , Curr. Opin. Struct. Biol. 52, 1 (2018). [OpenUrl][17][CrossRef][18] 2. [↵][19]1. Y. Cheng , Curr. Opin. Struct. Biol. 52, 58 (2018). [OpenUrl][20][CrossRef][21][PubMed][22] 3. [↵][23]1. M. Dong et al ., Nat. Commun. 11, 4137 (2020). [OpenUrl][24] 4. [↵][25]1. T. Nakane, 2. D. Kimanius, 3. E. Lindahl, 4. S. H. W. Scheres , eLife 7, e36861 (2018). [OpenUrl][26][CrossRef][27][PubMed][28] 5. [↵][29]1. W. Kühlbrandt , Science 343, 1443 (2014). [OpenUrl][30][Abstract/FREE Full Text][31] 6. [↵][32]1. Y. Cheng , Science 361, 876 (2018). [OpenUrl][33][Abstract/FREE Full Text][34] 7. [↵][35]1. K. Naydenova, 2. P. Jia, 3. C. J. Russo , Science 370, 223 (2020). [OpenUrl][36][CrossRef][37] 8. [↵][38]1. A. Douangamath et al ., bioRxiv 118117 (2020). 9. [↵][39]1. D. Lyumkis , J. Biol. Chem. 294, 5181 (2019). [OpenUrl][40][Abstract/FREE Full Text][41] 10. [↵][42]1. K. Naydenova et al ., IUCrJ 6, 1086 (2019). [OpenUrl][43][CrossRef][44][PubMed][45] 11. [↵][46]1. G. Scapin, 2. C. S. Potter, 3. B. Carragher , Cell Chem. Biol. 25, 1318 (2018). [OpenUrl][47] 12. [↵][48]1. T. Nakane et al ., bioRxiv 110189 (2020). 13. [↵][49]1. K. M. Yip, 2. N. Fischer, 3. E. Paknia, 4. A. Chari, 5. H. Stark , bioRxiv 106740 (2020). 14. [↵][50]1. D. Tegunov, 2. L. Xue, 3. C. Dienemann, 4. P. Cramer, 5. J. Mahamid , bioRxiv 136341 (2020). 15. [↵][51]1. A. Dance , Nat. Methods 17, 879 (2020). [OpenUrl][52] Acknowledgments: Supported by NIH grants GM103310 and OD019994 and Simons Foundation grant SF349247. [1]: #ref-1 [2]: #ref-2 [3]: #ref-3 [4]: #ref-4 [5]: #ref-5 [6]: #ref-6 [7]: #ref-7 [8]: #ref-8 [9]: #ref-9 [10]: #ref-10 [11]: #ref-11 [12]: #ref-12 [13]: #ref-13 [14]: #ref-14 [15]: #ref-15 [16]: #xref-ref-1-1 "View reference 1 in text" [17]: {openurl}?query=rft.jtitle%253DCurr.%2BOpin.%2BStruct.%2BBiol.%26rft.volume%253D52%26rft.spage%253D1%26rft_id%253Dinfo%253Adoi%252F10.1016%252Fj.sbi.2018.07.001%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [18]: /lookup/external-ref?access_num=10.1016/j.sbi.2018.07.001&link_type=DOI [19]: #xref-ref-2-1 "View reference 2 in text" [20]: {openurl}?query=rft.jtitle%253DCurr.%2BOpin.%2BStruct.%2BBiol.%26rft.volume%253D52%26rft.spage%253D58%26rft_id%253Dinfo%253Adoi%252F10.1016%252Fj.sbi.2018.08.008%26rft_id%253Dinfo%253Apmid%252F30219656%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [21]: /lookup/external-ref?access_num=10.1016/j.sbi.2018.08.008&link_type=DOI [22]: /lookup/external-ref?access_num=30219656&link_type=MED&atom=%2Fsci%2F370%2F6513%2F171.atom [23]: #xref-ref-3-1 "View reference 3 in text" [24]: {openurl}?query=rft.jtitle%253DNat.%2BCommun.%26rft.volume%253D11%26rft.spage%253D4137%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [25]: #xref-ref-4-1 "View reference 4 in text" [26]: {openurl}?query=rft.jtitle%253DeLife%26rft.volume%253D7%26rft.spage%253De36861%26rft_id%253Dinfo%253Adoi%252F10.7554%252FeLife.36861%26rft_id%253Dinfo%253Apmid%252F29856314%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [27]: /lookup/external-ref?access_num=10.7554/eLife.36861&link_type=DOI [28]: /lookup/external-ref?access_num=29856314&link_type=MED&atom=%2Fsci%2F370%2F6513%2F171.atom [29]: #xref-ref-5-1 "View reference 5 in text" [30]: {openurl}?query=rft.jtitle%253DScience%26rft.stitle%253DScience%26rft.aulast%253DKuhlbrandt%26rft.auinit1%253DW.%26rft.volume%253D343%26rft.issue%253D6178%26rft.spage%253D1443%26rft.epage%253D1444%26rft.atitle%253DThe%2BResolution%2BRevolution%26rft_id%253Dinfo%253Adoi%252F10.1126%252Fscience.1251652%26rft_id%253Dinfo%253Apmid%252F24675944%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [31]: /lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjEzOiIzNDMvNjE3OC8xNDQzIjtzOjQ6ImF0b20iO3M6MjI6Ii9zY2kvMzcwLzY1MTMvMTcxLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ== [32]: #xref-ref-6-1 "View reference 6 in text" [33]: {openurl}?query=rft.jtitle%253DScience%26rft.stitle%253DScience%26rft.aulast%253DCheng%26rft.auinit1%253DY.%26rft.volume%253D361%26rft.issue%253D6405%26rft.spage%253D876%26rft.epage%253D880%26rft.atitle%253DSingle-particle%2Bcryo-EM--How%2Bdid%2Bit%2Bget%2Bhere%2Band%2Bwhere%2Bwill%2Bit%2Bgo%26rft_id%253Dinfo%253Adoi%252F10.1126%252Fscience.aat4346%26rft_id%253Dinfo%253Apmid%252F30166484%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [34]: /lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjEyOiIzNjEvNjQwNS84NzYiO3M6NDoiYXRvbSI7czoyMjoiL3NjaS8zNzAvNjUxMy8xNzEuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9 [35]: #xref-ref-7-1 "View reference 7 in text" [36]: {openurl}?query=rft.jtitle%253DScience%26rft.stitle%253DScience%26rft.aulast%253DNaydenova%26rft.auinit1%253DK.%26rft.volume%253D370%26rft.issue%253D6513%26rft.spage%253D223%26rft.epage%253D226%26rft.atitle%253DCryo-EM%2Bwith%2Bsub-1%2BA%2Bspecimen%2Bmovement%26rft_id%253Dinfo%253Adoi%252F10.1126%252Fscience.abb7927%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [37]: /lookup/external-ref?access_num=10.1126/science.abb7927&link_type=DOI [38]: #xref-ref-8-1 "View reference 8 in text" [39]: #xref-ref-9-1 "View reference 9 in text" [40]: {openurl}?query=rft.jtitle%253DJ.%2BBiol.%2BChem.%26rft_id%253Dinfo%253Adoi%252F10.1074%252Fjbc.REV118.005602%26rft_id%253Dinfo%253Apmid%252F30804214%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [41]: /lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MzoiamJjIjtzOjU6InJlc2lkIjtzOjExOiIyOTQvMTMvNTE4MSI7czo0OiJhdG9tIjtzOjIyOiIvc2NpLzM3MC82NTEzLzE3MS5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30= [42]: #xref-ref-10-1 "View reference 10 in text" [43]: {openurl}?query=rft.jtitle%253DIUCrJ%26rft.volume%253D6%26rft.spage%253D1086%26rft_id%253Dinfo%253Adoi%252F10.1107%252FS2052252519012612%26rft_id%253Dinfo%253Apmid%252F31709064%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [44]: /lookup/external-ref?access_num=10.1107/S2052252519012612&link_type=DOI [45]: /lookup/external-ref?access_num=31709064&link_type=MED&atom=%2Fsci%2F370%2F6513%2F171.atom [46]: #xref-ref-11-1 "View reference 11 in text" [47]: {openurl}?query=rft.jtitle%253DCell%2BChem.%2BBiol.%26rft.volume%253D25%26rft.spage%253D1318%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [48]: #xref-ref-12-1 "View reference 12 in text" [49]: #xref-ref-13-1 "View reference 13 in text" [50]: #xref-ref-14-1 "View reference 14 in text" [51]: #xref-ref-15-1 "View reference 15 in text" [52]: {openurl}?query=rft.jtitle%253DNat.%2BMethods%26rft.volume%253D17%26rft.spage%253D879%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx


Cryo-EM with sub-1 A specimen movement

Science

Single-particle cryogenic electron microscopy (cryo-EM) has become a go-to technique for structural biologists. Although data-processing and reconstruction methods have improved, innovations in sample preparation and data collection are essential to reliably achieve high-resolution reconstructions while also reducing the amount of time required per structure. Naydenova et al. tackled the issue of electron beam–induced particle movement, a major source of information loss, by designing a gold sample support that prevents buckling of the extremely thin layer of ice in which the particles are suspended (see Perspective by Rapp and Carragher). The negligible particle displacement permits extrapolation to “zero exposure” structure factors, revealing features typically lost in cryo-EM structures. Far fewer particles per unit resolution are required, which greatly accelerates structure determination, especially at high resolution. Science , this issue p. [223][1]; see also p. [171][2] Most information loss in cryogenic electron microscopy (cryo-EM) stems from particle movement during imaging, which remains poorly understood. We show that this movement is caused by buckling and subsequent deformation of the suspended ice, with a threshold that depends directly on the shape of the frozen water layer set by the support foil. We describe a specimen support design that eliminates buckling and reduces electron beam–induced particle movement to less than 1 angstrom. The design allows precise foil tracking during imaging with high-speed detectors, thereby lessening demands on cryostage precision and stability. It includes a maximal density of holes, which increases throughput in automated cryo-EM without degrading data quality. Movement-free imaging allows extrapolation to a three-dimensional map of the specimen at zero electron exposure, before the onset of radiation damage. [1]: /lookup/doi/10.1126/science.abb7927 [2]: /lookup/doi/10.1126/science.abd8035