A Deep Learning Approach for Predicting FDA’s 510(k) Medical Device Recalls Using Device Citation Relationships

References

• Abbasi A , Li J , Adjeroh D , Abate M , Zheng W (2019) Don’t mention it? Analyzing user-generated content signals for early adverse event warnings. Inform. Systems Res. 30(3):1007–1028.Link, Google Scholar
• Abbasi A , Parsons J , Pant G , Sheng ORL , Sarker S (2024) Pathways for design research on artificial intelligence. Inform. Systems Res. 35(2):441–459.Link, Google Scholar
• Agarwal R , Gao G , DesRoches C , Jha AK (2010) Research commentary—The digital transformation of healthcare: Current status and the road ahead. Inform. Systems Res. 21(4):796–809.Link, Google Scholar
• Ahmad F , Abbasi A , Kitchens B , Adjeroh D , Zeng D (2022) Deep learning for adverse event detection from web search. IEEE Trans. Knowledge Data Engrg. 34(6):2681–2695.Google Scholar
• Ahsen ME , Ayvaci MUS , Raghunathan S (2019) When algorithmic predictions use human-generated data: A bias-aware classification algorithm for breast cancer diagnosis. Inform. Systems Res. 30(1):97–116.Link, Google Scholar
• Bahdanau D , Cho K , Bengio Y (2016) Neural machine translation by jointly learning to align and translate. Preprint, submitted May 19, http://arxiv.org/abs/1409.0473.Google Scholar
• Bardhan I , Oh J-h(C) , Zheng Z(E) , Kirksey K (2015) Predictive analytics for readmission of patients with congestive heart failure. Inform. Systems Res. 26(1):19–39.Link, Google Scholar
• Bardhan I , Kohli R , Oborn E , Mishra A , Tan CH , Tremblay MC , Sarker S (2025) Human-centric information systems research on the digital future of healthcare. Inform. Systems Res. 36(1):1–20.Link, Google Scholar
• Basu S , Hassenplug JC (2012) Patient access to medical devices—A comparison of U.S. and European review processes. New England J. Medicine 367(6):485–488.Crossref, Google Scholar
• Bates DW , Levine DM , Salmasian H , Syrowatka A , Shahian DM , Lipsitz S , Zebrowski JP , et al. (2023) The safety of inpatient health care. New England J. Medicine 388(2):142–153.Crossref, Google Scholar
• Beltagy I , Peters ME , Cohan A (2020) Longformer: The long-document transformer. Preprint, submitted December 2, http://arxiv.org/abs/2004.05150.Google Scholar
• Bollobás B (1998) Random graphs. Modern Graph Theory , Graduate Texts in Mathematics, vol. 184 (Springer, New York), 215–252.Crossref, Google Scholar
• Breiman L (2001) Random forests. Machine Learn. 45(1):5–32.Crossref, Google Scholar
• Calin O (2020) Deep Learning Architectures: A Mathematical Approach (Springer, Cham, Switzerland).Crossref, Google Scholar
• Cami A , Manzi S , Arnold A , Reis BY (2013) Pharmacointeraction network models predict unknown drug-drug interactions. PLoS One 8(4):e61468.Crossref, Google Scholar
• Cerqueira V , Torgo L , Mozetič I (2020) Evaluating time series forecasting models: An empirical study on performance estimation methods. Machine Learn. 109(11):1997–2028.Crossref, Google Scholar
• Chawla NV , Bowyer KW , Hall LO , Kegelmeyer WP (2002) SMOTE: Synthetic minority over-sampling technique. J. Artificial Intelligence Res. 16:321–357.Crossref, Google Scholar
• Chen T , Guestrin C (2016) XGBoost: A scalable tree boosting system. Proc. 22nd ACM SIGKDD Internat. Conf. Knowledge Discovery Data Mining (Association for Computing Machinery, New York), 785–794.Google Scholar
• Chen D , Lin Y , Li W , Li P , Zhou J , Sun X (2020) Measuring and relieving the over-smoothing problem for graph neural networks from the topological view. Proc. AAAI Conf. Artificial Intelligence 34(4):3438–3445.Crossref, Google Scholar
• Cho K , van Merrienboer B , Bahdanau D , Bengio Y (2014) On the properties of neural machine translation: Encoder-decoder approaches. Preprint, submitted October 7, http://arxiv.org/abs/1409.1259.Google Scholar
• Christensen J (2024) Philips reaches $1.1 billion settlement for CPAP machine lawsuits, admits no fault or liability. CNN (April 30), www.cnn.com/2024/04/30/health/philips-1-billion-cpap-settlement-agreement/index.html.Google Scholar
• Chung J , Gulcehre C , Cho K , Bengio Y (2014) Empirical evaluation of gated recurrent neural networks on sequence modeling. Preprint, submitted December 11, http://arxiv.org/abs/1412.3555.Google Scholar
• Darby JL , Ketchen DJ , Ball GP , Mukherjee U (2023) CEO stock ownership, recall timing, and stock market penalties. Manufacturing Service Oper. Management 25(5):1909–1930.Link, Google Scholar
• Davis J , Goadrich M (2006) The relationship between precision-recall and ROC curves. Proc. 23rd Internat. Conf. Machine Learn. (Association for Computing Machinery, New York), 233–240.Google Scholar
• Devlin J, et al. (2018) BERT: Pre-training of deep bidirectional transformers for language understanding. Preprint, submitted October 11, https://arxiv.org/abs/1810.04805.Google Scholar
• Dhruva SS , Redberg RF (2012) Medical device regulation: Time to improve performance. PLoS Medicine 9(7):e1001277.Crossref, Google Scholar
• Donahoe GF (2021) Estimates medical device spending in the United States. Retrieved May 1, www.advamed.org/wp-content/uploads/2021/12/Estimates-Medical-Device-Spending-United-States-Report-2021.pdf.Google Scholar
• Ensign LG (2016) Using text mining of FDA reports to inform early signal detection of cardiovascular lead recalls. PhD dissertation, University of Colorado Denver, Denver.Google Scholar
• Everhart AO , Karaca-Mandic P , Redberg RF , Ross JS , Dhruva SS (2025) Late adverse event reporting from medical device manufacturers to the US Food and Drug Administration: Cross sectional study. BMJ 388:e081518.Crossref, Google Scholar
• Everhart AO , Sen S , Stern AD , Zhu Y , Karaca-Mandic P (2023) Association between regulatory submission characteristics and recalls of medical devices receiving 510(k) clearance. J. Amer. Medical Assoc. 329(2):144–156.Crossref, Google Scholar
• Food and Drug Administration (2011) The 510(k) program: Evaluating substantial equivalence in premarket notifications. Retrieved May 1, www.fda.gov/regulatory-information/search-fda-guidance-documents/510k-program-evaluating-substantial-equivalence-premarket-notifications-510k.Google Scholar
• Food and Drug Administration (2020) Recalls, corrections and removals (devices). Retrieved May 1, https://www.fda.gov/medical-devices/postmarket-requirements-devices/recalls-corrections-and-removals-devices.Google Scholar
• Food and Drug Administration (2023a) Best practices for selecting a predicate device to support a premarket notification [510(k)] submission—Draft guidance for industry and Food and Drug Administration staff. Retrieved October 24, https://tinyurl.com/FDA-510k-bestpractices.Google Scholar
• Food and Drug Administration (2023b) De novo classification request. Retrieved May 1, https://www.fda.gov/medical-devices/premarket-submissions-selecting-and-preparing-correct-submission/de-novo-classification-request.Google Scholar
• Food and Drug Administration (2023c) 510(k) clearances. Retrieved May 1, www.fda.gov/medical-devices/device-approvals-denials-and-clearances/510k-clearances.Google Scholar
• Food and Drug Administration (2023d) 510(k) database. Retrieved May 1, www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm.Google Scholar
• Food and Drug Administration (2023e) Manufacturer and User Facility Device Experience (MAUDE). Retrieved May 1, www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfmaude/search.cfm.Google Scholar
• Food and Drug Administration (2023f) Medical device recalls. Retrieved May 1, www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfres/res.cfm.Google Scholar
• Food and Drug Administration (2023g) Premarket notification 510(k). Retrieved May 1, https://tinyurl.com/fda-510k.Google Scholar
• Food and Drug Administration (2024) MDSAP risk management procedure. Retrieved May 1, www.fda.gov/medical-devices/medical-device-single-audit-program-mdsap/mdsap-risk-management-procedure.Google Scholar
• Fuhr T , George K , Pai J (2013) The business case for medical device quality. Report, McKinsey Center for Government, McKinsey & Co., New York.Google Scholar
• Graham BL , Steenbruggen I , Miller MR , Barjaktarevic IZ , Cooper BG , Hall GL , Hallstrand TS , et al. (2019) Standardization of spirometry 2019 update. An official American Thoracic Society and European Respiratory Society technical statement. Amer. J. Respiratory Critical Care Medicine 200(8):e70–e88.Crossref, Google Scholar
• Guo T , Bardhan IR , Ding Y , Zhang S (2025) An explainable artificial intelligence approach using graph learning to predict intensive care unit length of stay. Inform. Systems Res. 36(3):1478–1501.Link, Google Scholar
• Gupta S , Georgiou A , Sen S , Simon K , Karaca-Mandic P (2021) US trends in COVID-19-associated hospitalization and mortality rates before and after reopening economies. JAMA Health Forum 2(6):e211262.Crossref, Google Scholar
• Hamilton W , Ying Z , Leskovec J (2017) Inductive representation learning on large graphs. Proc. 31st Internat. Conf. Neural Inform. Processing Systems (Long Beach, CA), 1025−1035.Google Scholar
• Hearst MA , Dumais ST , Osuna E , Platt J , Scholkopf B (1998) Support vector machines. IEEE Intelligent Systems Appl. 13(4):18–28.Crossref, Google Scholar
• Hochreiter S , Schmidhuber J (1997) Long short-term memory. Neural Comput. 9(8):1735–1780.Crossref, Google Scholar
• Hu MY , Zhang G(P) , Jiang CX , Patuwo BE (1999) A cross-validation analysis of neural network out-of-sample performance in exchange rate forecasting. Decision Sci. 30(1):197–216.Crossref, Google Scholar
• Huang PS , Chen SD , Smaragdis P , Hasegawa-Johnson M (2012) Singing-voice separation from monaural recordings using robust principal component analysis. IEEE ICASSP (Kyoto, Japan), 57–60.Google Scholar
• Institute of Medicine (2011) Medical Devices and the Public’s Health: The FDA 510(k) Clearance Process at 35 Years (National Academies Press, Washington, DC).Google Scholar
• Jewett C (2023) CPAP maker agrees to $479 million settlement over defects. New York Times (September 7), https://www.nytimes.com/2023/09/07/health/cpap-defect-recall-philips-respironics.html.Google Scholar
• Jia Z , Lin S , Ying R , You J , Leskovec J , Aiken A (2020) Redundancy-free computation for graph neural networks. Proc. 26th ACM SIGKDD Internat. Conf. Knowledge Discovery Data Mining (Association of Computing Machinery, New York), 997–1005.Google Scholar
• Kadakia KT , Beckman AL , Ross JS , Krumholz HM (2022) Renewing the call for reforms to medical device safety—The case of penumbra. JAMA Internal Medicine 182(1):59–65.Crossref, Google Scholar
• Kadakia KT , Dhruva SS , Caraballo C , Ross JS , Krumholz HM (2023) Use of recalled devices in new device authorizations under the US Food and Drug Administration’s 510(k) pathway and risk of subsequent recalls. J. Amer. Medical Assoc. 329(2):136–143.Crossref, Google Scholar
• Karaca-Mandic P , Sen S , Georgiou A , Zhu Y , Basu A (2020) Association of COVID-19-related hospital use and overall COVID-19 mortality in the USA. J. General Internal Medicine , ePub ahead of print August 19, https://doi.org/10.1007/s11606-020-06084-7.Crossref, Google Scholar
• Keras (2024) Keras: Deep learning for humans. Retrieved May 1, https://keras.io/.Google Scholar
• Kipf TN , Welling M (2017) Semi-supervised classification with graph convolutional networks. Preprint, submitted February 22, http://arxiv.org/abs/1609.02907.Google Scholar
• Kitaev N , Kaiser Ł , Levskaya A (2020) Reformer: The efficient transformer. Preprint, submitted February 18, http://arxiv.org/abs/2001.04451.Google Scholar
• Kitchens B , Claggett JL , Abbasi A (2024) Timely, granular, and actionable: Designing a social listening platform for Public Health 3.0. MIS Quart. 48(3):899–929.Crossref, Google Scholar
• Lalani C , Kunwar EM , Kinard M , Dhruva SS , Redberg RF (2021) Reporting of death in US Food and Drug Administration medical device adverse event reports in categories other than death. JAMA Internal Medicine 181(9):1217–1223.Crossref, Google Scholar
• LeCun Y , Bengio Y , Hinton G (2015) Deep learning. Nature 521(7553):436–444.Crossref, Google Scholar
• Lee DDK , Cheng ZZQ , Mao C , Manzoor E (2025) Guided diverse concept miner (GDCM): Uncovering relevant constructs for managerial insights from text. Inform. Systems Res. 36(1):370–393.Link, Google Scholar
• Lefkovich C , Rothenberg S (2023) Identification of predicate creep under the 510(k) process: A case study of a robotic surgical device. PLoS One 18(3):e0283442.Crossref, Google Scholar
• Lin T, Fu X, Chen F, Li L (2021) A novel approach for code smells detection based on deep learning. Chen B, Huang X, eds. Applied Cryptography in Computer and Communications. AC3 2021, Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, vol. 386 (Springer, Cham, Switzerland).Google Scholar
• Maisel W (2011) 510(k) premarket notification analysis of FDA recall data. Public Health Effectiveness of the FDA 510k Clearance Process: Measuring Postmarket Performance and Other Select Topics Workshop report, National Academies Press, Washington, DC.Google Scholar
• Menard S (2002) Applied Logistic Regression Analysis (SAGE, Thousands Oaks, CA).Crossref, Google Scholar
• Mohamed M , Mohamed N , Kim JG (2023) Advancements in wearable EEG technology for improved home-based sleep monitoring and assessment: A review. Biosensors 13(12):1019.Crossref, Google Scholar
• Mohammadi Aria M , Erten A , Yalcin O (2019) Technology advancements in blood coagulation measurements for point-of-care diagnostic testing. Frontiers Bioengineering Biotechnology 7:395.Crossref, Google Scholar
• Mukherjee UK , Sinha KK (2018) Product recall decisions in medical device supply chains: A big data analytic approach to evaluating judgment bias. Production Oper. Management 27(10):1816–1833.Crossref, Google Scholar
• National Science Foundation (2024) NSF I-Corps. Retrieved May 1, www.nsf.gov/funding/initiatives/i-corps.Google Scholar
• Padmanabhan B , Fang X , Sahoo N , Burton-Jones A (2022) Machine learning in information systems research. MIS Quart. 46(1):iii–xviii.Crossref, Google Scholar
• Prasetiyo B, Alamsyah MAM, Baroroh N (2021) Evaluation performance recall and F2 score of credit card fraud detection unbalanced dataset using SMOTE oversampling technique. J. Phys.: Conf. Ser. 1918:042002. Google Scholar
• Rai A (2017) Editor’s comments: Diversity of design science research. MIS Quart. 41(1):iii–xviii.Crossref, Google Scholar
• Rathi VK , Ross JS (2019) Modernizing the FDA’s 510(k) pathway. New England J. Medicine 381(20):1891–1893.Crossref, Google Scholar
• Rosh J , Bell CM , Urbach DR (2021) The 510(k) ancestry of transvaginal mesh: When the subject is not a predicate. JAMA Surgery 156(8):701–702.Crossref, Google Scholar
• Russek-Cohen E , Feldblyum T , Whitaker KB , Hojvat S (2011) FDA perspectives on diagnostic device clinical studies for respiratory infections. Clinical Infectious Diseases 52(Suppl 4):S305–S311.Crossref, Google Scholar
• Saito T , Rehmsmeier M (2015) The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLoS One 10(3):e0118432.Crossref, Google Scholar
• Sawaya J , Champlain A , Cohen J , Avram M (2021) Barriers to reporting: Limitations of the Maude database. Dermatologic Surgery 47(3):424–425.Crossref, Google Scholar
• Scikit-learn (2024) Scikit-learn: Machine learning in Python. Accessed November 28, 2025, https://scikit-learn.org/stable/.Google Scholar
• Sedgwick (2024) U.S. product recalls surge 11% in 2023 to hit seven-year high. Retrieved May 1, www.sedgwick.com/press-release/us-product-recalls-surge-in-2023-to-hit-seven-year-high/.Google Scholar
• Vaswani A , Shazeer N , Parmar N , Uszkoreit J , Jones L , Gomez AN , Kaiser, Polosukhin IŁ (2017) Attention is all you need. Proc. 31st Internat. Conf. Neural Inform. Processing Systems (Long Beach, CA), 5998–6008.Google Scholar
• Veličković P , Cucurull G , Casanova A , Romero A , Liò P , Bengio Y (2018) Graph attention networks. Preprint, submitted February 4, http://arxiv.org/abs/1710.10903.Google Scholar
• Vo CD , Jiang B , Azad TD , Crawford NR , Bydon A , Theodore N (2020) Robotic spine surgery: Current state in minimally invasive surgery. Global Spine J. 10(2):34S–40S.Crossref, Google Scholar
• Weiss M , Mohr H (2018) Patients shocked, burned by device touted to treat pain. Associated Press (November 26), https://apnews.com/article/38145385944248b5bf8d442579505258.Google Scholar
• Yang K , Lau RYK , Abbasi A (2023) Getting personal: A deep learning artifact for text-based measurement of personality. Inform. Systems Res. 34(1):194–222.Link, Google Scholar
• Yegnanarayana B (2009) Artificial Neural Networks (PHI Learning Private Limited, New Delhi, India).Google Scholar
• Yu B , Yin H , Zhu Z (2018) Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting. Proc. Twenty-Seventh Internat. Joint Conf. Artificial Intelligence (IJCAI), 3634–3640.Google Scholar
• Yu S , Chai Y , Chen H , Sherman SJ , Brown RA (2022) Wearable sensor-based chronic condition severity assessment: An adversarial attention-based deep multisource multitask learning approach. MIS Quart. 45(3):1355–1394.Crossref, Google Scholar
• Zargar N , Carr A (2018) The regulatory ancestral network of surgical meshes. PLoS One 13(6):e0197883.Crossref, Google Scholar
• Zhao L , Song Y , Zhang C , Liu Y , Wang P , Lin T , Deng M , Li H (2020) T-GCN: A temporal graph convolutional network for traffic prediction. IEEE Trans. Intelligent Transportation Systems 21(9):3848–3858.Crossref, Google Scholar
• Zhou T , Wang Y , Yan L , Tan Y (2023) Spoiled for choice? Personalized recommendation for healthcare decisions: A multiarmed bandit approach. Inform. Systems Res. 34(4):1493–1512.Link, Google Scholar
• Zhu Q , Du B , Yan P (2019) Multi-hop convolutions on weighted graphs. Preprint, submitted November 12, http://arxiv.org/abs/1911.04978.Google Scholar
• Zhu Y , Everhart A , Mandic PK , Sen S (2020) Using NLP to extract predicate history from medical device approvals. 2020 Internat. Conf. Inform. Systems ICIS (Association for Information Systems, Hyderabad, India).Google Scholar