FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3193669350> ?p ?o ?g. }

• W3193669350 endingPage "695" @default.
• W3193669350 startingPage "693" @default.
• W3193669350 abstract "Related article, p. 762 Related article, p. 762 In this issue of Kidney Medicine, Verma et al1Verma A. Vipul C.C. Waikar S.S. Kolachalama V.B. Machine learning applications in nephrology: a bibliometric analysis comparing kidney studies to other medicine subspecialities.Kidney Med. 2021; 3: 762-767Abstract Full Text Full Text PDF Scopus (1) Google Scholar highlight the underuse of machine learning as a research tool in the field of nephrology. Through a bibliometric search approach, they found that kidney research had the lowest number of articles using machine learning when compared with 4 other organ-specific research areas (brain, heart, liver, and lung). Additionally, among machine learning articles, the National Institute of Diabetes and Digestive and Kidney Diseases had the lowest number of acknowledgements as a funding source compared with 7 other National Institutes of Health institution sponsors. Organ-specific specialty journals were also found to have fewer articles featuring machine learning when compared with broad interdisciplinary journals. These findings highlight the importance of educating the nephrology community about the potential advantages (as well as inherent limitations) of artificial intelligence (AI) research tools. Although the scarcity of publications within the nephrology field using machine learning has already been pointed out,2Chan L. Vaid A. Nadkarni G.N. Applications of machine learning methods in kidney disease: hope or hype?.Curr Opin Nephrol Hypertens. 2020; 29: 319-326Crossref PubMed Scopus (3) Google Scholar,3Saez-Rodriguez J. Rinschen M.M. Floege J. Kramann R. Big science and big data in nephrology.Kidney Int. 2019; 95: 1326-1337Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar Verma et al1Verma A. Vipul C.C. Waikar S.S. Kolachalama V.B. Machine learning applications in nephrology: a bibliometric analysis comparing kidney studies to other medicine subspecialities.Kidney Med. 2021; 3: 762-767Abstract Full Text Full Text PDF Scopus (1) Google Scholar rigorously document this finding while also placing this into perspective with other organ-specific fields and highlighting the disparity of funding specifically directed toward machine learning research within nephrology. The article is very timely because of the increasing number of clinical trials and rapid advancement in the field of diagnostics and therapeutics, areas that might get a further boost with the integration of machine learning. AI refers to the development of algorithms that allow a computer to perceive, learn from data, react, make predictions, and act.4Niel O. Bastard P. Artificial intelligence in nephrology: core concepts, clinical applications, and perspectives.Am J Kidney Dis. 2019; 74: 803-810Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar Machine learning is one of the branches of AI and comprises a set of algorithms, such as convoluted neural networks, random forest, and deep learning, that have the ability to learn and improve from experience without having been explicitly programmed for a specific task.4Niel O. Bastard P. Artificial intelligence in nephrology: core concepts, clinical applications, and perspectives.Am J Kidney Dis. 2019; 74: 803-810Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar Convoluted neural networks are particularly efficient at analyzing data that are spatially or temporally dependent, such as images, explaining their popularity in radiology and pathology machine learning reseach.5Liu Y. Chen P.C. Krause J. Peng L. How to read articles that use machine learning: users' guides to the medical literature.JAMA. 2019; 322: 1806-1816Crossref PubMed Scopus (144) Google Scholar One of the advantages of deep learning is its ability to analyze big and complex data and yield predictions that in general tend to be similar to or outperform human experts. Humans become expert in their fields over many years after gathering and analyzing millions of data points in their brains. The number of data points that can be analyzed by machine learning is huge, and machine learning has a clear speed advantage over humans, offering the potential of advancing our understanding much faster. The requirement of a substantial amount of data for training of convoluted neural networks can be a limiting factor when dealing with rare diseases, as is the case with glomerular diseases for instance.4Niel O. Bastard P. Artificial intelligence in nephrology: core concepts, clinical applications, and perspectives.Am J Kidney Dis. 2019; 74: 803-810Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar The creation of large consortiums, such as the Nephrotic Syndrome Study Network (NEPTUNE), the Kidney Precision Medicine Project (KPMP), the Chronic Renal Insufficiency Cohort (CRIC), and the Cure Glomerulonephritis (CureGN) Study, is a potential way to circumvent this issue.4Niel O. Bastard P. Artificial intelligence in nephrology: core concepts, clinical applications, and perspectives.Am J Kidney Dis. 2019; 74: 803-810Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar Data obtained through these consortiums have already been used in machine learning research. For instance, digitalized kidney pathology slides from the NEPTUNE database have been used to train a convoluted neural network to segment normal kidney histologic structures, including glomeruli, distal and proximal tubules, and peritubular capillaries.6Jayapandian C.P. Chen Y. Janowczyk A.R. et al.Development and evaluation of deep learning-based segmentation of histologic structures in the kidney cortex with multiple histologic stains.Kidney Int. 2021; 99: 86-101Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar The assessment of pathologic features, such as interstitial fibrosis and tubular atrophy, has also been the subject of machine learning studies,7Ginley B. Jen K.Y. Han S.S. et al.Automated computational detection of interstitial fibrosis, tubular atrophy, and glomerulosclerosis.J Am Soc Nephrol. 2021; 32: 837-850Crossref Google Scholar including for prediction of kidney survival.8Kolachalama V.B. Singh P. Lin C.Q. et al.Association of pathological fibrosis with renal survival using deep neural networks.Kidney Int Rep. 2018; 3: 464-475Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar Other pathologic states for which convoluted neural networks have been successfully trained to grade and classify include diabetic nephropathy,9Ginley B. Lutnick B. Jen K.Y. et al.Computational segmentation and classification of diabetic glomerulosclerosis.J Am Soc Nephrol. 2019; 30: 1953-1967Crossref PubMed Scopus (58) Google Scholar lupus nephritis,10Ginley B, Jen K-Y, Rosenberg A, et al. Fully automated classification of glomerular lesions in lupus nephritis. Paper presented at: SPIE Medical Imaging. March 16, 2020; Houston, TX. Proc. SPIE 11320, Medical Imaging 2020: Digital Pathology, 113200Y.Google Scholar and interpretation of immunofluorescence images.11Ligabue G. Pollastri F. Fontana F. et al.Evaluation of the classification accuracy of the kidney biopsy direct immunofluorescence through convolutional neural networks.Clin J Am Soc Nephrol. 2020; 15: 1445-1454Crossref PubMed Scopus (12) Google Scholar In another example of how large disease consortiums could facilitate future machine learning research, data from the KPMP database are being used to generate and feed ontologies that could be used on AI/machine learning projects to link various -omics profiles to clinical features.12Ong E. Wang L.L. Schaub J. et al.Modelling kidney disease using ontology: insights from the Kidney Precision Medicine Project.Nat Rev Nephrol. 2020; 16: 686-696Crossref PubMed Scopus (12) Google Scholar One of the major advantages of the consortiums is the annotation and review of the digital images by experts from all over the world. The annotated images help provide the ground truth for the development of supervised AI algorithms. After validation of the algorithms, these annotated images can be made available in the public domain to help individual institutes develop their own algorithms. In most studies, the κ for kidney pathology parameters is very high. Machine learning and open-sourced reviewed annotated images may help improve the reproducibility of the findings and patient outcomes. Given their ability to analyze and integrate large amounts of complex data, machine learning tools are ideal to facilitate understanding of genotype-phenotype relationships in kidney diseases, which requires integration of digital nephropathology and radiology, transcriptomics, proteomics, metabolomics, and genome sequencing, as well as other data modalities such as electronic health record repositories.13Sealfon R.S.G. Mariani L.H. Kretzler M. Troyanskaya O.G. Machine learning, the kidney, and genotype-phenotype analysis.Kidney Int. 2020; 97: 1141-1149Abstract Full Text Full Text PDF PubMed Scopus (9) Google Scholar Within the transplant field, one promising initiative is the creation of the Banff Digital Pathology Working Group,14Farris A.B. Moghe I. Wu S. et al.Banff Digital Pathology Working Group: Going digital in transplant pathology.Am J Transplant. 2020; 20: 2392-2399Crossref PubMed Scopus (9) Google Scholar which should increase the availability and use of whole slide images of kidney transplant biopsies for future use in AI/machine learning research. Future potential applications of AI//machine learning in nephropathology include the assessment of transplant donor biopsies15Marsh J.N. Matlock M.K. Kudose S. et al.Deep learning global glomerulosclerosis in transplant kidney frozen sections.IEEE Trans Med Imaging. 2018; 37: 2718-2728Crossref PubMed Scopus (53) Google Scholar that are often read by non-nephropathologists with limited expertise in medical renal pathology, 3-dimensional reconstruction of kidney biopsy tissue,16Huo Y. Deng R. Liu Q. Fogo A.B. Yang H. AI applications in renal pathology.Kidney Int. 2021; 99: 1309-1320Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar and smartphone-based technologies to assist with adequacy evaluation of kidney core biopsies in real time, among others (reviewed by Huo et al16Huo Y. Deng R. Liu Q. Fogo A.B. Yang H. AI applications in renal pathology.Kidney Int. 2021; 99: 1309-1320Abstract Full Text Full Text PDF PubMed Scopus (5) Google Scholar). A few examples of nephrology-specific AI applications include prediction of the development of left ventricular dysfunction in patients with chronic kidney disease (CKD),17Attia Z.I. Kapa S. Lopez-Jimenez F. et al.Screening for cardiac contractile dysfunction using an artificial intelligence-enabled electrocardiogram.Nat Med. 2019; 25: 70-74Crossref PubMed Scopus (258) Google Scholar risk for developing progressive immunoglobulin A nephropathy,18Geddes C.C. Fox J.G. Allison M.E. Boulton-Jones J.M. Simpson K. An artificial neural network can select patients at high risk of developing progressive IgA nephropathy more accurately than experienced nephrologists.Nephrol Dial Transplant. 1998; 13: 67-71Crossref PubMed Scopus (40) Google Scholar dry weight assessment in maintenance dialysis patients,19Niel O. Bastard P. Boussard C. Hogan J. Kwon T. Deschenes G. Artificial intelligence outperforms experienced nephrologists to assess dry weight in pediatric patients on chronic hemodialysis.Pediatr Nephrol. 2018; 33: 1799-1803Crossref PubMed Scopus (14) Google Scholar acute kidney injury prediction in inpatients20Tomasev N. Glorot X. Rae J.W. et al.A clinically applicable approach to continuous prediction of future acute kidney injury.Nature. 2019; 572: 116-119Crossref PubMed Scopus (279) Google Scholar including in the intensive care unit setting,21Le S. Allen A. Calvert J. et al.Convolutional neural network model for intensive care unit acute kidney injury prediction.Kidney Int Rep. 2021; 6: 1289-1298Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar identification of pathogen-specific immune fingerprints in peritoneal dialysis patients,22Zhang J. Friberg I.M. Kift-Morgan A. et al.Machine-learning algorithms define pathogen-specific local immune fingerprints in peritoneal dialysis patients with bacterial infections.Kidney Int. 2017; 92: 179-191Abstract Full Text Full Text PDF PubMed Scopus (27) Google Scholar and even noninvasive high potassium detection through deep learning of electrocardiogram (ECG) patterns on a smartwatch.23Galloway C.D. Valys A.V. Petterson F.L. et al.Non-invasive detection of hyperkalemia with a smartphone electrocardiogram and artificial intelligence [abstract].J Am Coll Cardiol. 2018; 71: A272Crossref Google Scholar,24Galloway C.D. Valys A.V. Shreibati J.B. et al.Development and validation of a deep-learning model to screen for hyperkalemia from the electrocardiogram.JAMA Cardiol. 2019; 4: 428-436Crossref PubMed Scopus (76) Google Scholar Similar smartphone-based technology has also been used to detect atrial fibrillation, and it is currently US Food and Drug Administration (FDA) approved.25Topol E.J. High-performance medicine: the convergence of human and artificial intelligence.Nat Med. 2019; 25: 44-56Crossref PubMed Scopus (1173) Google Scholar A database of FDA-approved health care AI applications26Benjamens S. Dhunnoo P. Meskó B. The state of artificial intelligence-based FDA-approved medical devices and algorithms: an online database.NPJ Digital Med. 2020; 3: 118Crossref PubMed Scopus (110) Google Scholar includes 36 applications in radiology, 16 in cardiology, 6 in internal medicine, 5 in neurology, 3 in ophthalmology, 2 each in endocrinology and psychiatry, and 1 in pathology and urology,26Benjamens S. Dhunnoo P. Meskó B. The state of artificial intelligence-based FDA-approved medical devices and algorithms: an online database.NPJ Digital Med. 2020; 3: 118Crossref PubMed Scopus (110) Google Scholar,27FDA-approved A. I.-based algorithms.https://medicalfuturist.com/fda-approved-ai-based-algorithms/Google Scholar again highlighting the paucity of nephrology-specific AI research leading to the development of clinical applications. The single urology-related FDA-approved AI application consists of a smartphone-based urinalysis test kit to be used for at-home diagnosis of urinary tract infections.28Peck A.D. FDA approves smartphone-based urinalysis test kit for at-home use that matches quality of clinical laboratory tests. Laboratory Pathology Web site.https://www.darkdaily.com/2019/03/06/fda-approves-smartphone-based-urinalysis-test-kit-for-at-home-use-that-matches-quality-of-clinical-laboratory-tests/Date accessed: July 6, 2021Google Scholar Given the extensive interface between kidney and cardiovascular diseases, many AI approaches that are currently being considered for patients with cardiovascular diseases could also prove useful among the kidney patient community. These include the use of wearable devices to detect hemodynamic changes, including blood pressure levels through photoplethysmography, biomechanical sensors incorporated into clothing or shoes that could continuously measure cardiac output, lung fluid volume and weight, and tattoo-like sensors based on microfluidics for continuous hemodynamic monitoring.29Bayoumy K. Gaber M. Elshafeey A. et al.Smart wearable devices in cardiovascular care: where we are and how to move forward.Nat Rev Cardiol. 2021; : 1-19PubMed Google Scholar,30Krittanawong C. Rogers A.J. Johnson K.W. et al.Integration of novel monitoring devices with machine learning technology for scalable cardiovascular management.Nat Rev Cardiol. 2021; 18: 75-91Crossref PubMed Scopus (22) Google Scholar Due to the similarities between the retinal and kidney microcirculation, retinopathy has been proposed as a noninvasive biomarker of microvascular disorders in patients with CKD.31Farrah T.E. Dhillon B. Keane P.A. Webb D.J. Dhaun N. The eye, the kidney, and cardiovascular disease: old concepts, better tools, and new horizons.Kidney Int. 2020; 98: 323-342Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar Machine learning–based algorithms have been developed to assist in the diagnosis and classification of diabetic retinopathy32Gulshan V. Peng L. Coram M. et al.Development and validation of a deep learning algorithm for detection of diabetic retinopathy in retinal fundus photographs.JAMA. 2016; 316: 2402-2410Crossref PubMed Scopus (2685) Google Scholar; a similar algorithm could be tested in retinal images from patients with CKD to assist in disease severity stratification, risk for CKD progression,33Grunwald J.E. Pistilli M. Ying G.S. et al.Association between progression of retinopathy and concurrent progression of kidney disease: findings from the Chronic Renal Insufficiency Cohort (CRIC) Study.JAMA Ophthalmol. 2019; 137: 767-774Crossref PubMed Scopus (11) Google Scholar or the development of cardiovascular disease.34Grunwald J.E. Pistilli M. Ying G.S. et al.Progression of retinopathy and incidence of cardiovascular disease: findings from the Chronic Renal Insufficiency Cohort Study.Br J Ophthalmol. 2021; 105: 246-252Crossref PubMed Scopus (2) Google Scholar However, all these promising potential clinical applications can only reach the bedside after extensive research and validation, for which funding and motivation from the nephrology community to pursue those studies are essential. In this regard, Verma et al’s findings could be used to advocate for increased support from funding agencies into AI/machine learning–based kidney research. As suggested by Verma et al,1Verma A. Vipul C.C. Waikar S.S. Kolachalama V.B. Machine learning applications in nephrology: a bibliometric analysis comparing kidney studies to other medicine subspecialities.Kidney Med. 2021; 3: 762-767Abstract Full Text Full Text PDF Scopus (1) Google Scholar strategies to increase awareness and interest of nephrologists regarding machine learning could include introducing AI machine learning applications early in medical training to increase future physicians’ familiarity with these tools. One such successful example includes the integration of an iPad-based ECG platform into preclerkship physiology teaching of first-year medical students at the University of California, Irvine School of Medicine.35Frisch E.H. Greb A.C. Youm J.H. Wiechmann W.F. Greenberg M.L. Illustrating clinical relevance in the preclerkship medical school curriculum through active learning with KardiaMobile electrocardiography.Adv Physiol Educ. 2021; 45: 48-52Crossref PubMed Google Scholar The same platform is also FDA approved to record, store, and transfer single-channel ECG rhythms. Currently, machine learning is not an integral part of nephrology fellowship or internal medicine training curriculum. Physicians are not trained to use this technology in practice. The current path for AI/machine learning applications is that 2 people interested in technology will talk and will come up with projects of mutual interest. Nephrology can do better than this. To expedite the development of this field, we need to include machine learning curriculum in training and develop more collaborative opportunities. Another initiative that should improve the nephrology community awareness and understanding of AI and machine learning tools is a recent review series published by a leading kidney specialty journal,2Chan L. Vaid A. Nadkarni G.N. Applications of machine learning methods in kidney disease: hope or hype?.Curr Opin Nephrol Hypertens. 2020; 29: 319-326Crossref PubMed Scopus (3) Google Scholar as well as the incorporation of AI/machine learning sessions in nephrology meetings such as the American Society of Nephrology Kidney Week and the International Society of Nephrology. Critically, Verma et al highlight the gap between nephrology and other organ-based research using AI/machine learning, calling attention to a field that can help improve health moving forward if this gap is addressed. Clarissa Cassol, MD and Shree Sharma, MD. None. The authors declare that they have no relevant financial interests. Received August 18, 2021, in response to an invitation from the journal. Direct editorial input from the Editor-in-Chief. Accepted in revised form August 20, 2021. Machine Learning Applications in Nephrology: A Bibliometric Analysis Comparing Kidney Studies to Other Medicine SubspecialitiesKidney MedicineVol. 3Issue 5PreviewArtificial intelligence driven by machine learning algorithms is being increasingly employed for early detection, disease diagnosis, and clinical management. We explored the use of machine learning–driven advancements in kidney research compared with other organ-specific fields. Full-Text PDF Open Access" @default.
• W3193669350 created "2021-08-30" @default.
• W3193669350 creator A5051555467 @default.
• W3193669350 creator A5085530223 @default.
• W3193669350 date "2021-09-01" @default.
• W3193669350 modified "2023-10-16" @default.
• W3193669350 title "Nephrology Lagging Behind in Machine Learning Utilization" @default.
• W3193669350 cites W2140166463 @default.
• W3193669350 cites W2557738935 @default.
• W3193669350 cites W2597330433 @default.
• W3193669350 cites W2783839600 @default.
• W3193669350 cites W2792902933 @default.
• W3193669350 cites W2812840717 @default.
• W3193669350 cites W2901226889 @default.
• W3193669350 cites W2908201961 @default.
• W3193669350 cites W2919089713 @default.
• W3193669350 cites W2928542791 @default.
• W3193669350 cites W2944073789 @default.
• W3193669350 cites W2952527443 @default.
• W3193669350 cites W2964696298 @default.
• W3193669350 cites W2969528126 @default.
• W3193669350 cites W2971487518 @default.
• W3193669350 cites W2988716771 @default.
• W3193669350 cites W3007636570 @default.
• W3193669350 cites W3011545441 @default.
• W3193669350 cites W3012413413 @default.
• W3193669350 cites W3013433670 @default.
• W3193669350 cites W3015113267 @default.
• W3193669350 cites W3032898884 @default.
• W3193669350 cites W3080677331 @default.
• W3193669350 cites W3086667591 @default.
• W3193669350 cites W3087585143 @default.
• W3193669350 cites W3087619578 @default.
• W3193669350 cites W3092371431 @default.
• W3193669350 cites W3125668339 @default.
• W3193669350 cites W3126783352 @default.
• W3193669350 cites W3132601888 @default.
• W3193669350 cites W3135751641 @default.
• W3193669350 cites W3176619972 @default.
• W3193669350 doi "https://doi.org/10.1016/j.xkme.2021.08.004" @default.
• W3193669350 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/8515076" @default.
• W3193669350 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34693249" @default.
• W3193669350 hasPublicationYear "2021" @default.
• W3193669350 type Work @default.
• W3193669350 sameAs 3193669350 @default.
• W3193669350 citedByCount "1" @default.
• W3193669350 countsByYear W31936693502022 @default.
• W3193669350 crossrefType "journal-article" @default.
• W3193669350 hasAuthorship W3193669350A5051555467 @default.
• W3193669350 hasAuthorship W3193669350A5085530223 @default.
• W3193669350 hasBestOaLocation W31936693501 @default.
• W3193669350 hasConcept C126322002 @default.
• W3193669350 hasConcept C142724271 @default.
• W3193669350 hasConcept C2776962539 @default.
• W3193669350 hasConcept C41008148 @default.
• W3193669350 hasConcept C54847362 @default.
• W3193669350 hasConcept C71924100 @default.
• W3193669350 hasConceptScore W3193669350C126322002 @default.
• W3193669350 hasConceptScore W3193669350C142724271 @default.
• W3193669350 hasConceptScore W3193669350C2776962539 @default.
• W3193669350 hasConceptScore W3193669350C41008148 @default.
• W3193669350 hasConceptScore W3193669350C54847362 @default.
• W3193669350 hasConceptScore W3193669350C71924100 @default.
• W3193669350 hasIssue "5" @default.
• W3193669350 hasLocation W31936693501 @default.
• W3193669350 hasLocation W31936693502 @default.
• W3193669350 hasLocation W31936693503 @default.
• W3193669350 hasOpenAccess W3193669350 @default.
• W3193669350 hasPrimaryLocation W31936693501 @default.
• W3193669350 hasRelatedWork W2135562825 @default.
• W3193669350 hasRelatedWork W2354181859 @default.
• W3193669350 hasRelatedWork W2375646802 @default.
• W3193669350 hasRelatedWork W2472614530 @default.
• W3193669350 hasRelatedWork W2748952813 @default.
• W3193669350 hasRelatedWork W2757096358 @default.
• W3193669350 hasRelatedWork W2899084033 @default.
• W3193669350 hasRelatedWork W3145783610 @default.
• W3193669350 hasRelatedWork W3193669350 @default.
• W3193669350 hasRelatedWork W2584937992 @default.
• W3193669350 hasVolume "3" @default.
• W3193669350 isParatext "false" @default.
• W3193669350 isRetracted "false" @default.
• W3193669350 magId "3193669350" @default.
• W3193669350 workType "article" @default.