Goto

Collaborating Authors

Pennsylvania


New machine learning method allows hospitals to share patient data -- privately

#artificialintelligence

PHILADELPHIA - To answer medical questions that can be applied to a wide patient population, machine learning models rely on large, diverse datasets from a variety of institutions. However, health systems and hospitals are often resistant to sharing patient data, due to legal, privacy, and cultural challenges. An emerging technique called federated learning is a solution to this dilemma, according to a study published Tuesday in the journal Scientific Reports, led by senior author Spyridon Bakas, PhD, an instructor of Radiology and Pathology & Laboratory Medicine in the Perelman School of Medicine at the University of Pennsylvania. Federated learning -- an approach first implemented by Google for keyboards' autocorrect functionality -- trains an algorithm across multiple decentralized devices or servers holding local data samples, without exchanging them. While the approach could potentially be used to answer many different medical questions, Penn Medicine researchers have shown that federated learning is successful specifically in the context of brain imaging, by being able to analyze magnetic resonance imaging (MRI) scans of brain tumor patients and distinguish healthy brain tissue from cancerous regions.


10 Free Programming Courses by MIT, IBM, Google, Microsoft, and Apple

#artificialintelligence

You will learn about variables, conditional execution, repeated execution and how we use functions. Once a student completes this course, they will be ready to take more advanced programming courses. This course covers Python 3. 4. Programming for the Web with JavaScript Course by University of Pennsylvania The basics of how the World Wide Web allows browsers to send and retrieve web content; Web browser internals, the Document Object Model (DOM), and jQuery; How to create dynamic, interactive web pages using JavaScript; Techniques for creating data-driven websites using modern web technologies; Client-side JavaScript libraries and frameworks; Server-side JavaScript application architecture, middleware, HTTP, and RESTful API design 5. Python Basics for Data Science This Python course provides a beginner-friendly introduction to Python for Data Science. Practice through lab exercises, and you'll be ready to create your first Python scripts on your own! 6. Introduction to Computer Science and Programming Using Python An introduction to computer science as a tool to solve real-world analytical problems using Python 3.5.


New machine learning method allows hospitals to share patient data--privately

#artificialintelligence

To answer medical questions that can be applied to a wide patient population, machine learning models rely on large, diverse datasets from a variety of institutions. However, health systems and hospitals are often resistant to sharing patient data, due to legal, privacy, and cultural challenges. An emerging technique called federated learning is a solution to this dilemma, according to a study published Tuesday in the journal Scientific Reports, led by senior author Spyridon Bakas, Ph.D., an instructor of Radiology and Pathology & Laboratory Medicine in the Perelman School of Medicine at the University of Pennsylvania. Federated learning--an approach first implemented by Google for keyboards' autocorrect functionality--trains an algorithm across multiple decentralized devices or servers holding local data samples, without exchanging them. While the approach could potentially be used to answer many different medical questions, Penn Medicine researchers have shown that federated learning is successful specifically in the context of brain imaging, by being able to analyze magnetic resonance imaging (MRI) scans of brain tumor patients and distinguish healthy brain tissue from cancerous regions.


A new framework for understanding dynamic representations in networked neural systems

#artificialintelligence

Groups of neurons in the human brain produce patterns of activity that represent information about the stimuli that one is perceiving and then convey these patterns to different brain regions via nerve cell junctions known as synapses. So far, most neuroscience studies have focused on the two primary components of neuron information processing individually (i.e., the representation of stimuli in the form of neural activity and the transmission of this information in networks that model neural interactions), rather than exploring them together. A team of researchers at the University of Pennsylvania recently reviewed literature investigating each of these two components, in order to develop a holistic framework that better describes how groups of neurons process information. Their paper, published in Nature Neuroscience, introduces a holistic theoretical perspective that could inform future neuroscience research focusing on neural information processing. "In the past decade or so, neuroscientists have used more sophisticated tools to understand how the brain represents things that it sees or hears in its environment," Harang Ju and Danielle Bassett, the two researchers who carried out the study, told Medical Xpress.


Getting started with NLP using NLTK

#artificialintelligence

Well, wondering what is NLTK? the Natural Language Toolkit, or more commonly NLTK, is a suite of libraries and programs for symbolic and statistical natural language processing (NLP) for English written in the Python programming language. It was developed by Steven Bird and Edward Loper in the Department of Computer and Information Science at the University of Pennsylvania. It is necessary to convert the text to lower case as it is case sensitive. Tokenize sentences to get the tokens of the text i.e breaking the sentences into words.


AI Startup Lets Foresters See the Wood for the Trees

#artificialintelligence

AI startup Trefos is helping foresters see the wood for the trees. Using custom lidar and camera-mounted drones, the Philadelphia-based company collects data for high-resolution, 3D forest maps. These metrics allow government agencies and the forestry industry to estimate the volume of timber and biomass in an area of forest, as well as the amount of carbon stored in the trees. With this unprecedented detail, foresters can make more informed decisions when, for example, evaluating the need for controlled burns to clear biomass and reduce the risk of wildfires. "Forests are often very dense, with a very repetitive layout," said Steven Chen, founder and CEO of the startup, a member of the NVIDIA Inception program, which supports startups from product development to deployment. "We can use deep learning algorithms to detect trees, isolate them from the surrounding branches and vines, and use those as landmarks."


ICML 2020 Announces Test of Time Award

#artificialintelligence

Organizers of the 37th International Conference on Machine Learning (ICML) have announced this year's Test of Time award, which goes to a team from the California Institute of Technology, University of Pennsylvania, Saarland University. The ICML Test of Time award recognizes an ICML paper from ten years ago that has proven influential, with significant impacts in the field, "including both research and practice." Authors: Niranjan Srinivas, Andreas Krause, Sham Kakade, Matthias Seeger Institutions: California Institute of Technology, University of Pennsylvania, Saarland University Abstract: Many applications require optimizing an unknown, noisy function that is expensive to evaluate. We formalize this task as a multiarmed bandit problem, where the payoff function is either sampled from a Gaussian process (GP) or has low RKHS norm. We resolve the important open problem of deriving regret bounds for this setting, which imply novel convergence rates for GP optimization.


Tracing cell trajectories in a biofilm

Science

Born in 1881 on a farm in Pennsylvania, Alice C. Evans dedicated her life to studying bacteria in dairy products. Early in her career, Alice became convinced that most bacteria display multicellular behavior as part of their life cycles. At the time, the morphological changes observed in bacterial life cycles created confusion among scientists. In 1928, as the first female president of the American Society for Microbiology, Alice wrote to the scientific community: “When one-celled organisms grow in masses, … individual cells influence and protect one another.” She continued, “Bacteriologists need not feel chagrinned … to admit that… forms they have considered as different genera are but stages in the life cycle of one species” ([ 1 ][1]). Nearly 100 years later, on page 71 of this issue, Qin et al. ([ 2 ][2]) make a substantial leap forward in deciphering cell dynamics in biofilms—groups of microorganisms that adhere to a surface, and each other, by excreting matrix components. In the interim period, microbiologists have learned that many bacteria organize in groups. This allows bacterial cells to achieve collectively what individuals in isolation cannot, thus conferring a selective advantage on the individuals. Multicellular behaviors help cells to migrate ([ 3 ][3]), resist antibiotic treatments ([ 4 ][4]), and protect themselves from predators ([ 5 ][5]). In recent years, microbiologists have begun to unravel the mechanisms behind these multicellular behaviors, by studying single-cell gene expression, growth rate regulation, and cell-to-cell interactions ([ 6 ][6]–[ 9 ][7]), as well as by developing tools to investigate the morphology and growth of entire bacterial biofilms ([ 10 ][8], [ 11 ][9]). A multicellular aggregate starts with a single founder cell that grows into a mature biofilm. Despite substantial progress, scientists still lack a detailed understanding of how bacterial cells are programmed to build multicellular structures. Each cell makes individual decisions—whether to divide, move, excrete chemicals, exert forces, or express extracellular matrix components—in response to its local environment. In turn, the local environment is determined by the collective decisions of all of its cells, played out as a mosaic over time in a three-dimensional (3D) space. A primary challenge to unraveling the mystery of how cells are programmed to produce a mature functional biofilm is that researchers lack the experimental tools needed to study how the dynamics of individual cells drive biofilm formation and structure. ![Figure][10] The building of biofilms A fountain-like flow of bacterial cells drives biofilm expansion. CREDIT: V. ALTOUNIAN/ SCIENCE In their elegant study, Qin et al. developed dual-view light-sheet microscopy to reconstruct single-cell trajectories in 3D Vibrio cholerae biofilms initiated by a single founder cell. This method fluorescently labeled cellular puncta, giving isotropic single-cell resolution in the biofilm with much less photobleaching than that seen with previous methods. This advance allowed the authors to carry out simultaneous imaging of 10,000 V. cholerae cells for the 16 hours it takes for the biofilm to develop, with 3-min intervals between subsequent images. This frequent imaging made it possible to track the trajectories of micrometer-sized cells, giving an unprecedented view into the behaviors of individual cells as the biofilm developed (see the figure). The measurements revealed a qualitative transition in an individual cell's behavior, in which Brownian motion changes to ballistic motion as the biofilm develops. This transition corresponds to a new phase of collective growth, when the biofilm as a whole begins its vertical expansion away from the substrate. In this phase, cells displayed two types of trajectories. Some of the cells expanded ballistically outward, whereas others became trapped at the substrate. Overall, these trajectories gave rise to a collective fountain-like flow, which transported some cells to the biofilm front, while bypassing the cells trapped at the substrate. This fountain-like flow allowed for fast lateral expansion of the biofilm. Cell tracking allowed Qin et al. to precisely quantify the dynamics of various cells, while also assessing how these dynamics differ for mutant cells that overproduce matrix components. To interpret the results, the authors built a mathematical model for the mechanics of biofilm expansion, balancing growth with substrate friction. By modeling different surface frictions and comparing the predicted cell motion with the observed cell motion, Qin et al. were able to explain the observed behavior as long as friction between the cells and surface was a dominant effect. This study of V. cholerae offers an exciting insight into how collective behavior can arise from processes operating at the single-cell level. The mechanisms uncovered with a gram-negative bacterial species likely will be generalizable across other bacterial types. For example, the qualitative transitions in biofilm expansion observed in this study have analogs in other bacterial biofilms. With the gram-positive bacterium Bacillus subtilis , a qualitative change in colony expansion is triggered by a cellular bistable switch in which cells expressing flagella produce extracellular matrices ([ 12 ][11], [ 13 ][12]). Osmolarity associated with matrix production drives colony expansion ([ 14 ][13]). More broadly, this study demonstrates the great potential for advances in imaging technology and computer vision to help unravel how collective behavior arises from the activity of individual cells and their interactions. However, there is much more going on inside a biofilm that cannot yet be seen. More complete information would allow researchers to not only reconstruct the motion of cells but also uncover their phenotypic states. Previous work on B. subtilis with fluorescent labeling of genetic components shows detailed spatial arrangement of various cell types, with cells carrying out different biological functions in distinct parts of the biofilm ([ 3 ][3], [ 15 ][14]). One can only hypothesize about the diversity of cellular types and functions inside the beautiful fountain revealed in the present study. A deeper understanding of bacterial multicellular behavior will increase our ability to treat bacterial infections, control natural bacterial communities, and engineer synthetic ones for specific purposes. 1. [↵][15]1. A. C. Evans , J. Bacteriol. 17, 63 (1929). [OpenUrl][16][FREE Full Text][17] 2. [↵][18]1. B. Qin et al ., Science 369, 71 (2020). [OpenUrl][19][Abstract/FREE Full Text][20] 3. [↵][21]1. J. van Gestel et al ., PLOS Biol. 13, e1002141 (2015). [OpenUrl][22][CrossRef][23][PubMed][24] 4. [↵][25]1. C. W. Hall, 2. T.-F. Mah , FEMS Microbiol. Rev. 41, 276 (2017). [OpenUrl][26][CrossRef][27] 5. [↵][28]1. P. K. Raghupathi et al ., Front. Microbiol. 8, 2649 (2018). [OpenUrl][29] 6. [↵][30]1. A. Dal Co, 2. S. van Vliet, 3. M. Ackermann , Philos. Trans. R. Soc. London Ser. B 374, 20190080 (2019). [OpenUrl][31] 7. 1. A. Dal Co et al ., Nat. Ecol. Evol. 4, 366 (2020). [OpenUrl][32] 8. 1. S. van Vliet et al ., Cell Syst. 6, 496 (2018). [OpenUrl][33] 9. [↵][34]1. A. Dragoš et al ., Curr. Biol. 28, 1903 (2018). [OpenUrl][35][CrossRef][36] 10. [↵][37]1. K. Drescher et al ., Proc. Natl. Acad. Sci. U.S.A. 113, E2066 (2016). [OpenUrl][38][Abstract/FREE Full Text][39] 11. [↵][40]1. R. Hartmann et al ., Nat. Phys. 15, 251 (2019). [OpenUrl][41][CrossRef][42][PubMed][43] 12. [↵][44]1. H. Vlamakis et al ., Chemtracts 20, 427 (2007). [OpenUrl][45] 13. [↵][46]1. D. B. Kearns et al ., Mol. Microbiol. 55, 739 (2005). [OpenUrl][47][CrossRef][48][PubMed][49][Web of Science][50] 14. [↵][51]1. A. Seminara et al ., Proc. Natl. Acad. Sci. U.S.A. 109, 1116 (2012). [OpenUrl][52][Abstract/FREE Full Text][53] 15. [↵][54]1. H. Vlamakis et al ., Nat. Rev. Microbiol. 11, 157 (2013). [OpenUrl][55][CrossRef][56][PubMed][57] Acknowledgments: A.D.C. and M.P.B. are supported by the National Science Foundation (DMS-1715477), Materials Research Science and Engineering Center (DMR-1420570), the Office of Naval Research (N00014-17-1-3029), and the Simons Foundation. [1]: #ref-1 [2]: #ref-2 [3]: #ref-3 [4]: #ref-4 [5]: #ref-5 [6]: #ref-6 [7]: #ref-9 [8]: #ref-10 [9]: #ref-11 [10]: pending:yes [11]: #ref-12 [12]: #ref-13 [13]: #ref-14 [14]: #ref-15 [15]: #xref-ref-1-1 "View reference 1 in text" [16]: {openurl}?query=rft.jtitle%253DJournal%2Bof%2BBacteriology%26rft.stitle%253DJ.%2BBacteriol.%26rft.aulast%253DEvans%26rft.auinit1%253DA.%2BC.%26rft.volume%253D17%26rft.issue%253D2%26rft.spage%253D63%26rft.epage%253D77%26rft.atitle%253DLIFE%2BCYCLES%2BIN%2BBACTERIA.%26rft_id%253Dinfo%253Apmid%252F16559356%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 [17]: /lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6MzoiUERGIjtzOjExOiJqb3VybmFsQ29kZSI7czoyOiJqYiI7czo1OiJyZXNpZCI7czo3OiIxNy8yLzYzIjtzOjQ6ImF0b20iO3M6MjE6Ii9zY2kvMzY5LzY0OTkvMzAuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9 [18]: #xref-ref-2-1 "View reference 2 in text" [19]: {openurl}?query=rft.jtitle%253DScience%26rft.stitle%253DScience%26rft.aulast%253DQin%26rft.auinit1%253DB.%26rft.volume%253D369%26rft.issue%253D6499%26rft.spage%253D71%26rft.epage%253D77%26rft.atitle%253DCell%2Bposition%2Bfates%2Band%2Bcollective%2Bfountain%2Bflow%2Bin%2Bbacterial%2Bbiofilms%2Brevealed%2Bby%2Blight-sheet%2Bmicroscopy%26rft_id%253Dinfo%253Adoi%252F10.1126%252Fscience.abb8501%26rft_id%253Dinfo%253Apmid%252F32527924%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 [20]: /lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6Mzoic2NpIjtzOjU6InJlc2lkIjtzOjExOiIzNjkvNjQ5OS83MSI7czo0OiJhdG9tIjtzOjIxOiIvc2NpLzM2OS82NDk5LzMwLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ== [21]: #xref-ref-3-1 "View reference 3 in text" [22]: {openurl}?query=rft.jtitle%253DPLOS%2BBiol.%26rft.volume%253D13%26rft.spage%253De1002141%26rft_id%253Dinfo%253Adoi%252F10.1371%252Fjournal.pbio.1002141%26rft_id%253Dinfo%253Apmid%252F25894589%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 [23]: /lookup/external-ref?access_num=10.1371/journal.pbio.1002141&link_type=DOI [24]: /lookup/external-ref?access_num=25894589&link_type=MED&atom=%2Fsci%2F369%2F6499%2F30.atom [25]: #xref-ref-4-1 "View reference 4 in text" [26]: {openurl}?query=rft.jtitle%253DFEMS%2BMicrobiol.%2BRev.%26rft.volume%253D41%26rft.spage%253D276%26rft_id%253Dinfo%253Adoi%252F10.1093%252Ffemsre%252Ffux010%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=10.1093/femsre/fux010&link_type=DOI [28]: #xref-ref-5-1 "View reference 5 in text" [29]: {openurl}?query=rft.jtitle%253DFront.%2BMicrobiol.%26rft.volume%253D8%26rft.spage%253D2649%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 [30]: #xref-ref-6-1 "View reference 6 in text" [31]: {openurl}?query=rft.jtitle%253DPhilos.%2BTrans.%2BR.%2BSoc.%2BLondon%2BSer.%2BB%26rft.volume%253D374%26rft.spage%253D20190080%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 [32]: {openurl}?query=rft.jtitle%253DNat.%2BEcol.%2BEvol.%26rft.volume%253D4%26rft.spage%253D366%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 [33]: {openurl}?query=rft.jtitle%253DCell%2BSyst.%26rft.volume%253D6%26rft.spage%253D496%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]: #xref-ref-9-1 "View reference 9 in text" [35]: {openurl}?query=rft.jtitle%253DCurr.%2BBiol.%26rft.volume%253D28%26rft.spage%253D1903%26rft_id%253Dinfo%253Adoi%252F10.1016%252Fj.cub.2018.04.046%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 [36]: /lookup/external-ref?access_num=10.1016/j.cub.2018.04.046&link_type=DOI [37]: #xref-ref-10-1 "View reference 10 in text" [38]: {openurl}?query=rft.jtitle%253DProc.%2BNatl.%2BAcad.%2BSci.%2BU.S.A.%26rft_id%253Dinfo%253Adoi%252F10.1073%252Fpnas.1601702113%26rft_id%253Dinfo%253Apmid%252F26933214%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 [39]: /lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoicG5hcyI7czo1OiJyZXNpZCI7czoxMjoiMTEzLzE0L0UyMDY2IjtzOjQ6ImF0b20iO3M6MjE6Ii9zY2kvMzY5LzY0OTkvMzAuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9 [40]: #xref-ref-11-1 "View reference 11 in text" [41]: {openurl}?query=rft.jtitle%253DNat.%2BPhys.%26rft.volume%253D15%26rft.spage%253D251%26rft_id%253Dinfo%253Adoi%252F10.1038%252Fs41567-018-0356-9%26rft_id%253Dinfo%253Apmid%252F31156716%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 [42]: /lookup/external-ref?access_num=10.1038/s41567-018-0356-9&link_type=DOI [43]: /lookup/external-ref?access_num=31156716&link_type=MED&atom=%2Fsci%2F369%2F6499%2F30.atom [44]: #xref-ref-12-1 "View reference 12 in text" [45]: {openurl}?query=rft.jtitle%253DChemtracts%26rft.volume%253D20%26rft.spage%253D427%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 [46]: #xref-ref-13-1 "View reference 13 in text" [47]: {openurl}?query=rft.jtitle%253DMolecular%2Bmicrobiology%26rft.stitle%253DMol%2BMicrobiol%26rft.aulast%253DKearns%26rft.auinit1%253DD.%2BB.%26rft.volume%253D55%26rft.issue%253D3%26rft.spage%253D739%26rft.epage%253D749%26rft.atitle%253DA%2Bmaster%2Bregulator%2Bfor%2Bbiofilm%2Bformation%2Bby%2BBacillus%2Bsubtilis.%26rft_id%253Dinfo%253Adoi%252F10.1111%252Fj.1365-2958.2004.04440.x%26rft_id%253Dinfo%253Apmid%252F15661000%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]: /lookup/external-ref?access_num=10.1111/j.1365-2958.2004.04440.x&link_type=DOI [49]: /lookup/external-ref?access_num=15661000&link_type=MED&atom=%2Fsci%2F369%2F6499%2F30.atom [50]: /lookup/external-ref?access_num=000226457800008&link_type=ISI [51]: #xref-ref-14-1 "View reference 14 in text" [52]: {openurl}?query=rft.jtitle%253DProc.%2BNatl.%2BAcad.%2BSci.%2BU.S.A.%26rft_id%253Dinfo%253Adoi%252F10.1073%252Fpnas.1109261108%26rft_id%253Dinfo%253Apmid%252F22232655%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 [53]: /lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoicG5hcyI7czo1OiJyZXNpZCI7czoxMDoiMTA5LzQvMTExNiI7czo0OiJhdG9tIjtzOjIxOiIvc2NpLzM2OS82NDk5LzMwLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ== [54]: #xref-ref-15-1 "View reference 15 in text" [55]: {openurl}?query=rft.jtitle%253DNat.%2BRev.%2BMicrobiol.%26rft.volume%253D11%26rft.spage%253D157%26rft_id%253Dinfo%253Adoi%252F10.1038%252Fnrmicro2960%26rft_id%253Dinfo%253Apmid%252F23353768%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 [56]: /lookup/external-ref?access_num=10.1038/nrmicro2960&link_type=DOI [57]: /lookup/external-ref?access_num=23353768&link_type=MED&atom=%2Fsci%2F369%2F6499%2F30.atom


Radiology: Artificial Intelligence

#artificialintelligence

Radiology: Artificial Intelligence will host its second tweet chat on July 1, 2020, from 8:00 to 9:00 pm Eastern Daylight Time (U.S.), on the topic of interpretability of AI algorithms in radiology. The tweet chat will be moderated by Dr. Despina Kontos, deputy editor of this journal and associate professor of radiology at the University of Pennsylvania, and Dr. Aimilia Gastounioti, a research associate in the department of radiology at the University of Pennsylvania. The article discusses the methods that allow AI systems to explain their decisions through visualization, counterexamples, and semantics--and the many challenges to bring interpretability methods into clinical practice. Enhancing interpretability is essential to allow for AI systems to be trusted and verified for faster and more reliable adoption into clinical workflows. They explore whether patients and radiologists can better trust a model that explains its decisions, and how interpretability may accelerate the translation of deep learning tools into clinical practice.


Is Deep Learning Necessary For Simple Classification Tasks

#artificialintelligence

Deep learning (DL) models are known for tackling the nonlinearities associated with data, which the traditional estimators such as logistic regression couldn't. However, there is still a cloud of doubt with regards to the increased use of computationally intensive DL for simple classification tasks. To find out if DL really outperforms shallow models significantly, the researchers from the University of Pennsylvania experiment with three ML pipelines that involve traditional methods, AutoML and DL in a paper titled, 'Is Deep Learning Necessary For Simple Classification Tasks.' The UPenn researchers stated that a support-vector machine (SVM) model might predict more accurately susceptibility to a certain complex genetic disease than a gradient boosting model trained on the same dataset. Moreover, choosing different hyperparameters within that SVM model can vary performances.