Pith. sign in

REVIEW 5 major objections 5 minor 56 references

Advancing clinical trial outcomes using deep learning and predictive modelling: bridging precision medicine and patient-centered care

T0 review · 5 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper reports that an integrated pipeline of CNNs, RNNs, and transformers can cut clinical trial recruitment time by 25% and operational costs by 30% while predicting outcomes at 88–93% accuracy.

desk verdict A review-style paper whose headline results are unsupported by any artifact; the survey skeleton is fine as background reading, but the empirical claims should not survive review. read the letter →

arxiv 2412.07050 v1 pith:IOB2YR35 submitted 2024-12-09 cs.LG

Sydney Anuyah , Mallika K Singh , Hope Nyavor This is my paper · ORCID
classification cs.LG
keywords DeeplearningPredictivemodellingClinicaltrialsPrecisionmedicineNaturallanguageprocessingPatientstratificationMultimodaldataAdverseeventprediction
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to show that deep learning can address the biggest practical failures of clinical trials: slow recruitment, high cost, and avoidable adverse events. It reports training convolutional, recurrent, and transformer networks on a custom multimodal dataset of structured clinical records, genomic profiles, demographic data, and unstructured text, and claims the resulting pipeline reduced recruitment time by 25%, cut operational costs by 30%, and achieved predictive scores of 92% accuracy and 0.96 ROC-AUC for the CNN, 88% F1 for the RNN, and 93% precision for the transformer. The importance, if the results hold, is that trial design could become adaptive and patient-specific rather than based on population averages, which would shorten drug development timelines and make trials safer and more inclusive. The paper also surveys the surrounding literature and frames the approach as a general integration framework for precision medicine.

What carries the argument

The load-bearing mechanism is a multimodal fusion pipeline in which a CNN processes imaging data, an RNN/LSTM processes temporal clinical data, and a transformer processes unstructured text, with outputs combined through ensemble learning and dynamic risk scoring. The custom dataset, described as combining 50,000 structured patient records, 10,000 genomic records, 30,000 demographic records, and 20,000 unstructured clinical notes alongside GAN-generated synthetic data, is what carries the reported metrics; without it the architecture comparison has no empirical grounding.

What would settle it

A reader could reconstruct an equivalent multimodal dataset from public critical-care records, genomic databases, demographic surveys, and clinical notes, run CNN, RNN/LSTM, and transformer models under the stated 70/15/15 stratified split, and check whether the reported figures (CNN 92% accuracy and 0.96 ROC-AUC, RNN 88% F1, transformer 93% precision, 25% faster recruitment, 30% lower costs) reproduce on the held-out test set. Non-reproduction within normal statistical variance would falsify the paper's quantitative central claim.

Watch

Extended reading notes

Core claim

The paper's central claim is that an integrated deep-learning pipeline — CNNs for imaging, LSTMs for sequential vital-sign data, transformers for clinical text, plus survival analysis and dynamic risk scoring — can be trained on a fused multimodal dataset and outperform traditional statistical methods across the clinical trial workflow. The reported numbers are the evidence: 92% accuracy and 0.96 ROC-AUC for tumor detection in imaging, 88% F1 for adverse-event prediction from time series, 93% precision for extracting insights from trial protocols and notes, a 25% reduction in patient recruitment time, a 30% reduction in operational costs, and simulated trial success rates of 80% for the transformer pipeline versus 60% for traditional methods. The authors present these results as demonstrating that predictive analytics can be integrated into precision medicine to streamline trial design, monitoring, and patient-centered care.

Load-bearing premise

The load-bearing assumption is that the custom multimodal dataset described in the paper — 50,000 structured clinical records, 10,000 genomic records, 30,000 demographic records, and 20,000 unstructured text records — was actually assembled, cleaned, and used to train the models, because every reported performance and efficiency figure depends on it and the paper provides no dataset release, code, or access procedure.

Editorial extensions

If this is right

  • If the reported recruitment-speed gain translates to real trials, the 80% of studies that currently miss enrollment deadlines could be brought back on schedule.
  • Real-time RNN monitoring at the claimed recall level would let trial teams intervene before adverse events become serious, changing the safety monitoring workflow.
  • Transformer-based protocol and note analysis at 93% precision could automate much of the manual data-extraction and regulatory-documentation burden in trials.
  • Adaptive designs driven by these predictions would let trial sponsors re-randomize or adjust protocols as evidence accumulates, rather than waiting for trial end.
  • The 30% cost reduction, if real, would mean tens of millions of dollars saved per drug development program given the $2.5 billion average cost cited in the paper.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not describe an external validation cohort or a pre-registered analysis, so the reported numbers should be read as in-sample results unless independent replication appears.
  • A fair comparison would require the same data, the same train/test split, and the same hyperparameters to be run by another group; until then, the 25% and 30% efficiency gains are plausibility arguments rather than measured effects.
  • The framework's clinical usefulness would be tested by a prospective trial in which recruitment time and adverse-event rates are measured against a concurrent control arm, not against historical baselines.
  • If the dataset were released with schema and preprocessing code, the three architectures could be re-benchmarked side by side, which would also let researchers weigh the transformer's higher compute cost against its precision advantage.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 5 minor

Summary. The paper claims to train and evaluate CNN, RNN/LSTM, and transformer/BERT models on a custom multimodal clinical dataset, reporting high accuracy (CNN 92%), ROC-AUC (0.96), F1 (RNN 88%), precision (transformer 93%), and operational gains such as 25% faster recruitment and 30% lower costs. It also presents use cases for patient stratification, adverse event prediction, and personalized medicine. The manuscript describes data sources, preprocessing, model architectures, and an experimental setup, but provides no dataset, code, hyperparameter details, split sizes, baseline comparisons, or raw result tables. Every reported metric and operational claim depends on an unverifiable custom dataset described in Section 3.2 and Table 1.

Significance. If the reported results were substantiated, the paper could offer a useful demonstration of deep learning applied to clinical trial workflows, and the proposed framework of integrating imaging, temporal, and text data is a plausible direction. However, the contribution is currently not assessable: there is no reproducible experiment, no dataset release, no code, no statistical uncertainty quantification, and no comparison against established baselines. The paper also does not provide a machine-checked derivation or any verifiable artifact. The central empirical claims are therefore unsupported, and the scientific value cannot be confirmed from the manuscript as submitted.

major comments (5)
  1. [Section 3.2, Table 1] The entire experimental section depends on a custom dataset described in Section 3.2 and Table 1, but the dataset is not available and its description is internally inconsistent. Table 1 lists 10,000 'genomic records' from the 1000 Genomes Project, yet the public release of that project contains 2,504 individuals; the manuscript does not define what a 'record' means here, and no accession or processing details are given. Similarly, the 50,000 MIMIC-III records, 30,000 demographics records, and 20,000 text records are asserted without a data dictionary, inclusion/exclusion operationalization, or evidence that these data were actually assembled and merged. Because every metric in Section 4 is derived from this dataset, the reported results are unverifiable.
  2. [Section 3.2 and Section 4.1.1] The manuscript states that synthetic GAN-generated data were added to augment real-world data, but Section 4 does not state what fraction of the test set is synthetic or whether the reported accuracy, precision, recall, and ROC-AUC values were computed on real, synthetic, or mixed records. If the metrics are partly or wholly based on generated data, the clinical interpretation of the results is invalid. The paper must either report separate results on real and synthetic test sets or justify why synthetic data are valid for clinical evaluation. This missing information is load-bearing for the central claim that the models achieve the reported performance.
  3. [Section 4.1.1 and Table 4] The model performance numbers (CNN 92% accuracy, 0.96 ROC-AUC; RNN 88% F1; transformer 93% precision) are presented as experimental results, but the paper provides no training details, no hyperparameter values, no exact train/validation/test split sizes, no confidence intervals or error bars, and no baseline comparisons to standard methods. In particular, Section 3.6.1 states a 70/15/15 split but does not report the actual number of samples in each set, and no classifier comparison is made to logistic regression, random forests, or other standard approaches. As written, the numbers in Table 4 are narrative assertions rather than reproducible measurements.
  4. [Section 4.1.3 and Figure 9] The operational claims — 25% reduction in recruitment time, 30% reduction in manual data processing costs, up to $500,000 savings per trial, and trial success rates of 80% for transformers versus 65% for RNNs and 60% for traditional methods — are not tied to any experiment, simulation, or statistical analysis described in the methodology. No data, model, or procedure is provided for how these percentages were obtained, and the cited references do not support these specific numbers. These claims are load-bearing for the paper's conclusion that deep learning materially improves clinical trial workflows, yet they are unsupported.
  5. [Section 3.7 and Section 4] The case studies in Section 3.7 — for example, 85% glioblastoma stratification accuracy, 92% cardiotoxicity prediction accuracy, and a 20% reduction in hypoglycemic episodes — are introduced with 'Example:' and appear to be illustrative, but the Results section then presents similar numbers as experimental outcomes without clarifying which values are measured and which are illustrative. This conflation makes it impossible to determine which results were actually obtained by the authors versus which are hypothetical or borrowed from other studies. The manuscript must clearly distinguish measured results from illustrative scenarios, and it must provide supporting data for any measured result.
minor comments (5)
  1. [Title page] The copyright line contains a typo: 'Liscense' should be 'License'.
  2. [References [15] and [51]] Reference [15] and reference [51] both list 'Attention is all you need' with the same authors, and reference [51] is malformed; this duplicate should be consolidated and properly formatted.
  3. [Figure 4 caption] The caption 'Visualization of Case studies with visualizations of stratified patient groups' is redundant; it should be simplified, for example to 'Visualization of stratified patient groups using t-SNE plots.'
  4. [Section 3.5.4] The text says 'PyTorch:' with a trailing colon instead of a period, and the formatting of the tool list is inconsistent; this should be corrected for readability.
  5. [References] Several references are unrelated to the statements they are attached to, such as reference [7] on sustainable packaging attached to a claim about IBM Watson for Clinical Trials, and references [19]–[21] attached to data preprocessing claims; the authors should revise the citation list so that each claim is supported by a relevant source.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper reports asserted experimental numbers rather than deriving predictions from fitted inputs, and its self-citations are not load-bearing.

full rationale

The paper reports experimental metrics for CNN, RNN, and transformer models on a custom multimodal clinical dataset, but it contains no formal derivation chain in which a predicted quantity is shown to equal a fitted input by construction. The Section 3.2 dataset description and Table 1 are unverified and internally inconsistent (e.g., 10,000 genomic records vs. the 1000 Genomes Project's 2,504 samples, and a CNN imaging use case despite no imaging records appearing in Table 1), and no code or data are released. These are reproducibility and correctness risks, not circularity: the reported numbers are asserted rather than derived, and no equation or fitted parameter is renamed as a prediction. The paper's self-citations ([45], [47], and [54], which involve author Anuyah) appear in general, illustrative, or context-setting passages and do not carry the load-bearing claim that the models achieve 92% accuracy, 0.96 ROC-AUC, 25% faster recruitment, or 30% lower cost. There is no uniqueness theorem, no imported ansatz, no renaming of a known result, and no prediction that reduces by construction to its own input. Accordingly, the circularity score is 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

No explicit model weights or fitted parameters are disclosed. The reported accuracies and impact percentages function as claimed measurements, but without code or data they cannot be attributed to any concrete fitted model. The paper introduces no new particles, forces, dimensions, or mediators. Its unsupported elements are empirical claims, not invented entities.

assumptions (3)
  • domain assumption A custom dataset with the record counts in Table 1 exists and is suitable for the described tasks.
    Section 3.2 and Table 1 describe 50,000 MIMIC-III trial records, 10,000 genomic records, 30,000 demographic records, and 20,000 unstructured text records, but no dataset is released or linked.
  • domain assumption The reported performance metrics in Section 4 reflect actual held-out evaluations of the described models.
    Section 4.1.1 gives accuracy, precision, recall, and ROC-AUC values without code, evaluation scripts, or error analysis artifacts.
  • ad hoc to paper The case-study examples (e.g., 85% glioblastoma stratification, 92% cardiotoxicity prediction, 20% reduction in hypoglycemic episodes) are genuine observed results rather than illustrative scenarios.
    Section 3.7 and Section 6.1 present these as concrete outcomes, but no data source, patient cohort, or reproducible workflow is supplied.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Advancing clinical trial outcomes using deep learning and predictive modelling: bridging precision medicine and patient-centered care." pith.science (2026). https://pith.science/paper/IOB2YR35

@misc{pith2026241207050,
  author       = {Pith},
  title        = {Pith review of: Advancing clinical trial outcomes using deep learning and predictive modelling: bridging precision medicine and patient-centered care},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IOB2YR35}},
  note         = {Machine review of arXiv:2412.07050}
}
read the original abstract

The integration of artificial intelligence [AI] into clinical trials has revolutionized the process of drug development and personalized medicine. Among these advancements, deep learning and predictive modelling have emerged as transformative tools for optimizing clinical trial design, patient recruitment, and real-time monitoring. This study explores the application of deep learning techniques, such as convolutional neural networks [CNNs] and transformerbased models, to stratify patients, forecast adverse events, and personalize treatment plans. Furthermore, predictive modelling approaches, including survival analysis and time-series forecasting, are employed to predict trial outcomes, enhancing efficiency and reducing trial failure rates. To address challenges in analysing unstructured clinical data, such as patient notes and trial protocols, natural language processing [NLP] techniques are utilized for extracting actionable insights. A custom dataset comprising structured patient demographics, genomic data, and unstructured text is curated for training and validating these models. Key metrics, including precision, recall, and F1 scores, are used to evaluate model performance, while trade-offs between accuracy and computational efficiency are examined to identify the optimal model for clinical deployment. This research underscores the potential of AI-driven methods to streamline clinical trial workflows, improve patient-centric outcomes, and reduce costs associated with trial inefficiencies. The findings provide a robust framework for integrating predictive analytics into precision medicine, paving the way for more adaptive and efficient clinical trials. By bridging the gap between technological innovation and real-world applications, this study contributes to advancing the role of AI in healthcare, particularly in fostering personalized care and improving overall trial success rates.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

56 extracted references · 51 canonical work pages

  1. [1]

    2021 Global life sciences outlook

    Deloitte. 2021 Global life sciences outlook. Available from: https://www2.deloitte.com/global/en/insights

    work page 2021

  2. [2]

    Clinical trials: Challenges and opportunities

    Gul RB, Ali PA. Clinical trials: Challenges and opportunities. JPMA. 2010;60[6]:485 –489. Available from: https://jpma.org.pk/

    work page 2010

  3. [3]

    Barriers to participation in clinical trials

    Mills EJ, Seely D, Rachlis B, et al. Barriers to participation in clinical trials. Lancet Oncology. 2006;7[2]:141–148. doi:10.1016/S1470-2045[06]70576-9

    work page doi:10.1016/s1470-2045 2006

  4. [4]

    Epidemiology of chronic myeloid leukaemia: an update

    Höglund M, Sandin F, Simonsson B. Epidemiology of chronic myeloid leukaemia: an update. Annals of hematology. 2015 Apr;94:241-7

    work page 2015

  5. [5]

    The creative destruction of medicine: How the digital revolution will create better health care

    Topol EJ. The creative destruction of medicine: How the digital revolution will create better health care. 2nd ed. New York: Basic Books; 2020

    work page 2020

  6. [6]

    The role of artificial intelligence in improving healthcare outcomes

    McKinsey & Company. The role of artificial intelligence in improving healthcare outcomes. Available from: https://www.mckinsey.com/ai-healthcare

    work page

  7. [7]

    The role of sustainable packaging in enhancing brand loyalty among climate-conscious consumers in fast-moving consumer goods [FMCG]

    Adesoye A. The role of sustainable packaging in enhancing brand loyalty among climate-conscious consumers in fast-moving consumer goods [FMCG]. Int Res J Mod Eng Technol Sci. 2024;6[3]:112 -130. doi:10.56726/IRJMETS63233

    work page doi:10.56726/irjmets63233 2024

  8. [8]

    Use of Artificial Intelligence in Improving Outcomes in Heart Disease: A Scientific Statement From the American Heart Association

    Armoundas AA, Narayan SM, Arnett DK, Spector -Bagdady K, Bennett DA, Celi LA, Friedman PA, Gollob MH, Hall JL, Kwitek AE, Lett E. Use of Artificial Intelligence in Improving Outcomes in Heart Disease: A Scientific Statement From the American Heart Association. Circulation. 2024 Apr 2;149[14]:e1028-50

    work page 2024

Show all 56 references
  1. [9]

    Adaptive design methods in clinical trials

    Chow SC, Chang M. Adaptive design methods in clinical trials. Biostatistics. 2008;9[2]:293 –297. doi:10.1093/biostatistics/kxm040

    2008 doi

  2. [10]

    AI-powered chatbots in clinical trials: A systematic review

    Verma A, Sarangi S, Pal RK. AI-powered chatbots in clinical trials: A systematic review. Journal of Clinical Research. 2020;12[3]:245–256. doi:10.1016/j.jcr.2020.05.003

    2020 doi

  3. [11]

    Patient -centred care in clinical trials

    World Health Organization [WHO]. Patient -centred care in clinical trials. Available from: https://www.who.int/clinical-trials

  4. [12]

    Next -generation cancer organoids

    LeSavage BL, Suhar RA, Broguiere N, Lutolf MP, Heilshorn SC. Next -generation cancer organoids. Nature materials. 2022 Feb;21[2]:143-59

    2022

  5. [13]

    Revolutionizing Lassa fever prevention: Cutting - edge MATLAB image processing for non-invasive disease control [Internet]

    Chukwunweike JN, Dolapo H, Yussuf, Adewale MF and Victor I. Revolutionizing Lassa fever prevention: Cutting - edge MATLAB image processing for non-invasive disease control [Internet]. Vol. 23, World Journal of Advanced Research and Reviews. GSC Online Press; 2024. Available fr...

    2024 doi

  6. [14]

    Ethical and clinical implications of AI integration in cardiovascular healthcare

    Ahmad A, Husnain A, Shiwlani A, Hussain A, Gondal MN, Saeed A. Ethical and clinical implications of AI integration in cardiovascular healthcare. World Journal of Advanced Research and Reviews. 2024;23[3]:2479-501

    2024

  7. [15]

    Attention is all you need

    Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need. In: Advances in Neural Information Processing Systems; 2017. p. 5998–6008

    2017

  8. [16]

    Survival of patients with chronic myelocytic leukemia: comparisons of estimates from clinical trial settings and population-based cancer registries

    Pulte D, Gondos A, Redaniel MT, Brenner H. Survival of patients with chronic myelocytic leukemia: comparisons of estimates from clinical trial settings and population-based cancer registries. The Oncologist. 2011 May 1;16[5]:663-71

    2011

  9. [17]

    Predictive Analytics for Cyber Threat Intelligence in Fintech Using Big Data and Machine Learning

    Ekundayo F, Atoyebi I, Soyele A, Ogunwobi E. Predictive Analytics for Cyber Threat Intelligence in Fintech Using Big Data and Machine Learning. Int J Res Publ Rev . 2024;5[11]:1 -15. Available from: https://ijrpr.com/uploads/V5ISSUE11/IJRPR35463.pdf

    2024

  10. [18]

    Joseph Nnaemeka Chukwunweike, Moshood Yussuf, Oluwatobiloba Okusi, Temitope Oluwatobi Bakare, Ayokunle J. Abisola. The role of deep learning in ensuring privacy integrity and security: Applications in AI-driven World Journal of Advanced Research and Reviews, 2024, 24(03), 001–...

    2024

  11. [19]

    The second machine age: Work, progress, and prosperity in a time of brilliant technologies

    Brynjolfsson E, McAfee A. The second machine age: Work, progress, and prosperity in a time of brilliant technologies. New York: Norton & Company; 2014

    2014

  12. [20]

    Novel Innovation Design for the Future of Health: Entrepreneurial Concepts for Patient Empowerment and Health Democratization

    Friebe M, editor. Novel Innovation Design for the Future of Health: Entrepreneurial Concepts for Patient Empowerment and Health Democratization. Springer Nature; 2022 Nov 26

    2022

  13. [21]

    AI in healthcare: Opportunities and barriers

    Ellen MacArthur Foundation. AI in healthcare: Opportunities and barriers. Available from: https://www.ellenmacarthurfoundation.org/ai-healthcare

  14. [22]

    Radiopharmaceutical therapy in cancer: clinical advances and challenges

    Sgouros G, Bodei L, McDevitt MR, Nedrow JR. Radiopharmaceutical therapy in cancer: clinical advances and challenges. Nature reviews Drug discovery. 2020 Sep 1;19[9]:589-608

    2020

  15. [23]

    MIMIC -III, a freely accessible critical care database

    Johnson AE, Pollard TJ, Shen L, et al. MIMIC -III, a freely accessible critical care database. Scientific Data . 2016;3[1]:160035. doi:10.1038/sdata.2016.35

    2016 doi

  16. [24]

    A global reference for human genetic variation

    Auton A, Abecasis GR, Altshuler DM, et al. A global reference for human genetic variation. Nature. 2015;526[7571]:68–74. doi:10.1038/nature15393

    2015 doi

  17. [25]

    Generating multi -label discrete patient records using generative adversarial networks

    Choi E, Biswal S, Malin B, et al. Generating multi -label discrete patient records using generative adversarial networks. Proceedings of the 2nd Machine Learning for Healthcare Conference. 2017. p. 286–305

    2017

  18. [26]

    Data anonymization practices in clinical trials

    European Medicines Agency. Data anonymization practices in clinical trials. Available from: https://www.ema.europa.eu/

  19. [27]

    Visualizing data using t -SNE

    Van der Maaten L, Hinton G. Visualizing data using t -SNE. Journal of Machine Learning Research . 2008;9[11]:2579–2605

    2008

  20. [28]

    Advancing Precision Healthcare: The Integration of Nanotechnology, Millimeter Wave Sensing, Laser Technology, Fibre Bragg Grating, and Deep Learning Models

    Omenogor, Christian E. and Adewale Abayomi Adeniran. “Advancing Precision Healthcare: The Integration of Nanotechnology, Millimeter Wave Sensing, Laser Technology, Fibre Bragg Grating, and Deep Learning Models.” International Journal of Research Publication and Reviews [2024]:...

    2024

  21. [29]

    Applications of machine learning in drug discovery and development

    Vamathevan J, Clark D, Czodrowski P, Dunham I, Ferran E, Lee G, Li B, Madabhushi A, Shah P, Spitzer M, Zhao S. Applications of machine learning in drug discovery and development. Nature reviews Drug discovery. 2019 Jun;18[6]:463-77

    2019

  22. [30]

    Sample size requirements to estimate key design parameters from external pilot randomised controlled trials: a simulation study

    Teare MD, Dimairo M, Shephard N, Hayman A, Whitehead A, Walters SJ. Sample size requirements to estimate key design parameters from external pilot randomised controlled trials: a simulation study. Trials. 2014 Dec;15:1-3

    2014

  23. [31]

    Leveraging AI -Driven Decision Intelligence for Complex Systems Engineering

    Ekundayo F. Leveraging AI -Driven Decision Intelligence for Complex Systems Engineering. Int J Res Publ Rev . 2024;5[11]:1-10. Available from: https://ijrpr.com/uploads/V5ISSUE11/IJRPR35397.pdf

    2024

  24. [32]

    Leveraging AI and machine learning for the protection of critical national infrastructure

    Okusi O. Leveraging AI and machine learning for the protection of critical national infrastructure. Asian Journal of Research in Computer Science. 2024 Sep 27;17[10]:1-1. http://dx.doi.org/10.9734/ajrcos/2024/v17i10505

    2024 doi

  25. [33]

    Joseph Nnaemeka Chukwunweike, Abayomi Adejumo. Leveraging AI and Principal Component Analysis [PCA] For In-Depth Analysis in Drilling Engineering: Optimizing Production Metrics through Well Logs and Reservoir Data https://dx.doi.org/10.7753/ijcatr1309.1004

    arXiv

  26. [34]

    CRISPR in cancer biology and therapy

    Katti A, Diaz BJ, Caragine CM, Sanjana NE, Dow LE. CRISPR in cancer biology and therapy. Nature Reviews Cancer. 2022 May;22[5]:259-79

    2022

  27. [35]

    Predictive Modelling of Loop Execution and Failure Rates in Deep Learning Systems: An Advanced MATLAB Approach https://www.doi.org/10.56726/IRJMETS61029

    Chukwunweike JN, Kayode Blessing Adebayo, Moshood Yussuf, Chikwado Cyril Eze, Pelumi Oladokun, Chukwuemeka Nwachukwu. Predictive Modelling of Loop Execution and Failure Rates in Deep Learning Systems: An Advanced MATLAB Approach https://www.doi.org/10.56726/IRJMETS61029

    doi

  28. [36]

    Factors affecting success of new drug clinical trials

    Kim E, Yang J, Park S, Shin K. Factors affecting success of new drug clinical trials. Therapeutic Innovation & Regulatory Science. 2023 Jul;57[4]:737-50

    2023

  29. [37]

    Sustainable economic growth through artificial intelligence -driven tax frameworks nexus on enhancing business efficiency and prosperity: An appraisal

    Shallon Asiimire, Baton Rouge, Fechi George Odocha, Friday Anwansedo, Oluwaseun Rafiu Adesanya. Sustainable economic growth through artificial intelligence -driven tax frameworks nexus on enhancing business efficiency and prosperity: An appraisal. International Journal of Late...

    2024

  30. [38]

    Artificial intelligence to deep learning: machine intelligence approach for drug discovery

    Gupta R, Srivastava D, Sahu M, Tiwari S, Ambasta RK, Kumar P. Artificial intelligence to deep learning: machine intelligence approach for drug discovery. Molecular diversity. 2021 Aug;25:1315-60

    2021

  31. [39]

    AI -driven anomaly detection for proactive cybersecurity and data breach prevention

    Nwoye CC, Nwagwughiagwu S. AI -driven anomaly detection for proactive cybersecurity and data breach prevention. Zenodo; 2024. Available from: https://doi.org/10.5281/zenodo.14197924

    2024 doi

  32. [40]

    Estimating clinical trial success rates and related parameters in oncology

    Wong CH, Siah KW, Lo AW. Estimating clinical trial success rates and related parameters in oncology. Available at SSRN 3355022. 2019 Mar 19

    2019

  33. [41]

    Joseph Nnaemeka Chukwunweike, Moshood Yussuf , Oluwatobiloba Okusi, Temitope Oluwatobi Bakare and Ayokunle J. Abisola. The role of deep learning in ensuring privacy integrity and security:Applications in AI-driven cybersecurity solutions https://dx.doi.org/10.30574/wjarr.2024....

    doi

  34. [42]

    Cellular senescence and senolytics: the path to the clinic

    Chaib S, Tchkonia T, Kirkland JL. Cellular senescence and senolytics: the path to the clinic. Nature medicine. 2022 Aug;28[8]:1556-68

    2022

  35. [43]

    Success probability estimation with applications to clinical trials

    De Martini D. Success probability estimation with applications to clinical trials. John Wiley & Sons; 2013 May 6

    2013

  36. [44]

    Harnessing digital platforms for sustainable marketing: strategies to reduce single -use plastics in consumer behaviour

    Adesoye A. Harnessing digital platforms for sustainable marketing: strategies to reduce single -use plastics in consumer behaviour. Int J Res Publ Rev. 2024;5[11]:44-63. doi:10.55248/gengpi.5.1124.3102

    2024

  37. [45]

    Data-to-question generation using deep learning

    Koshy NR, Dixit A, Jadhav SS, Penmatsa AV, Samanthapudi SV, Kumar MGA, Anuyah SO, Vemula G, Herzog PS, Bolchini D. Data-to-question generation using deep learning. In: 2023 4th International Conference on Big Data Analytics and Practices [IBDAP]. IEEE; 2023. p. 1–6

    2023

  38. [46]

    The role of artificial intelligence in improving patient outcomes and future of healthcare delivery in cardiology: a narrative review of the literature

    Gala D, Behl H, Shah M, Makaryus AN. The role of artificial intelligence in improving patient outcomes and future of healthcare delivery in cardiology: a narrative review of the literature. InHealthcare 2024 Feb 16 [Vol. 12, No. 4, p. 481]. MDPI

    2024

  39. [47]

    Can deep learning large language models be used to unravel knowledge graph creation? In: Proceedings of the International Conference on Computing, Machine Learning and Data Science

    Anuyah S, Chakraborty S. Can deep learning large language models be used to unravel knowledge graph creation? In: Proceedings of the International Conference on Computing, Machine Learning and Data Science. 2024. p. 1–6

    2024

  40. [48]

    A survey on deep learning in medical image analysis

    Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis. Medical Image Analysis. 2017;42:60–88. doi:10.1016/j.media.2017.07.005

    2017 doi

  41. [49]

    Artificial intelligence for clinical trial design

    Harrer S, Shah P, Antony B, Hu J. Artificial intelligence for clinical trial design. Trends in pharmacological sciences. 2019 Aug 1;40[8]:577-91

    2019

  42. [50]

    Estimation of clinical trial success rates and related parameters

    Wong CH, Siah KW, Lo AW. Estimation of clinical trial success rates and related parameters. Biostatistics. 2019;20[2]:273–286. doi:10.1093/biostatistics/kxx069

    2019 doi

  43. [51]

    In: Advances in Neural Information Processing Systems; 2017

    Vaswani A, Shazeer N, Parmar N, Attention is all you need. In: Advances in Neural Information Processing Systems; 2017. p. 5998–6008

    2017

  44. [52]

    Predictive analytics in clinical trials

    McKinsey & Company. Predictive analytics in clinical trials. Available from: https://www.mckinsey.com/predictive-clinical-trials

  45. [53]

    AI-driven solutions for clinical trials

    IBM Watson. AI-driven solutions for clinical trials. Available from: https://www.ibm.com/watson-health

  46. [54]

    Understanding graph databases: a comprehensive tutorial and survey

    Anuyah S, Bolade V, Agbaakin O. Understanding graph databases: a comprehensive tutorial and survey. arXiv preprint arXiv:2411.09999 [Internet]. 2024 [cited YYYY -MM-DD]; Available from: https://arxiv.org/abs/2411.09999

    2024 arXiv

  47. [55]

    Artificial intelligence in health care: The hope, the hype, the promise, the peril

    Matheny M, Israni ST, Ahmed M, et al. Artificial intelligence in health care: The hope, the hype, the promise, the peril. NAM Special Publication. 2019. Available from: https://nam.edu/ai-in-healthcare

    2019

  48. [56]

    AI/ML Action Plan for Software as a Medical Device

    FDA. AI/ML Action Plan for Software as a Medical Device. Available from: https://www.fda.gov/medical-devices

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.