Artificial Intelligence in Oncology: Current Landscape, Challenges, and Future Directions

In the paper

“A search of ClinicalTrials.gov of”

In the paper

“This includes the storage of patient data in electronic medical record (EMR) systems, the digitization of radiology and pathology imaging (3), and the increasing adoption of routine genomic profiling (4).”

What the sources say

• Not checked yetcites [3] Niazi, 2019, Digital pathology and artificial intelligence, Lancet Oncol source
• Not checked yetcites [4] AACR Project GENIE Consortium, 2017, AACR Project GENIE: powering precision medicine through an international consortium, Cancer Discov source

In the paper

“the review focuses on the four most common cancer types: breast, prostate, lung, and colorectal, accounting for 50% of all new cancer cases in 2023 (7).”

What the sources say

• Not checked yetcites [7] Surveillance, Epidemiology, and End Results (SEER): cancer stat facts: common cancer sites

In the paper

“Artificial intelligence (AI) is increasingly being applied to all aspects of oncology.”

In the paper

“In particular, advances in deep learning (1) over the past decade have made it feasible to learn complex patterns directly from real-world data, making it the core driver of AI advances in and out of healthcare (2).”

What the sources say

• Not checked yetcites [1] LeCun, 2015, Deep learning, Nature source
• Not checked yetcites [2] Topol, 2019, High-performance medicine: the convergence of human and artificial intelligence, Nat Med source

In the paper

“In the case of mammography, there are now multiple commercial AI devices cleared by the U.S. Federal Drug Administration (FDA) and/or the European Union (EU) that are designed to aid radiologists in breast cancer detection.”

In the paper

“Although not yet uniformly adopted across imaging providers or cancer hospitals, there are multiple FDAcleared and/or EU-certified commercial devices in the cancer detection domain and multiple reported clinical trials.”

In the paper

“There are also multiple ongoing clinical trials evaluating the real-world effectiveness of such systems.”

In the paper

“the Mirai system developed by Yala and colleagues ( 12) is designed to predict 5-year cancer risk directly from mammograms and has been retrospectively evaluated across multiple hospitals.”

What the sources say

• Not checked yetcites [12] Yala, 2022, Multi-institutional validation of a mammography-based breast cancer risk model, J Clin Oncol source

In the paper

“These data can be used to build and train AI models, and critically, the real-time availability of these data can enable individualized, clinically relevant AI predictions to further the goal of precision oncology (5).”

What the sources say

• Not checked yetcites [5] Garraway, 2013, Precision oncology: an overview, J Clin Oncol source

In the paper

“AI applications for these cancer types serve as a good representation of the current state of the field, especially as their prevalence facilitates large-scale data collection and encourages clinical application.”

In the paper

“a longer-term, prospective clinical trial is now seeking to assess the connection between colonoscopy CADe systems and incidence and mortality from colorectal cancer after a 10-year follow-up period (25).”

What the sources say

• Not checked yetcites [25] Mori, 2022, Clinical trial designs for artificial intelligence in gastrointestinal endoscopy, Lancet Gastroenterol Hepatol source

In the paper

“groups have reported on the use of deep learning for breast cancer diagnosis and subtyping from H&E slides (49)(50)(51), and quantifying other clinically relevant features such as mitotic cell counting (52,53).”

What the sources say

• Not checked yetcites [49] Sui, 2021, A pyramid architecture-based deep learning framework for breast cancer detection, BioMed Res Int source
• Not checked yetcites [50] Sandbank, 2022, Validation and real-world clinical application of an artificial intelligence algorithm for breast cancer detection in biopsies, NPJ Breast Cancer source
• Not checked yetcites [51] Couture, 2018, Image analysis with deep learning to predict breast cancer grade, ER status, histologic subtype, and intrinsic subtype, NPJ Breast Cancer source

In the paper

“The CAMELYON16/17 data have subsequently become a standard benchmark for deep learning models in pathology.”

In the paper

“In a retrospective study, pathologists demonstrated improved sensitivity in diagnosing prostate cancer when using the AI platform, along with a reduction in review time and no statistically significant change in specificity.”

What the sources say

• Not checked yetcites [60] Raciti, 2020, Novel artificial intelligence system increases the detection of prostate cancer in whole slide images of core needle biopsies, Mod Pathol source

In the paper

“Since 2013, the U.S. Preventive Services Task Force has recommended low-dose computed tomography to screen for lung cancer in high-risk populations (26).”

What the sources say

• Not checked yetcites [26] US Preventive Services Task Force, 2021, Screening for lung cancer: US Preventive Services Task Force Recommendation Statement, JAMA source

In the paper

“The AI company Paige has also recently been granted FDA Breakthrough Device Designation status for Paige Lymph Node, providing a path for accelerated FDA clearance (48).”

What the sources say

• Not checked yetcites [48] U.S. FDA grants Paige breakthrough device designation for cancer detection in breast lymph nodes

In the paper

“There have been many AI efforts in this domain, including those aimed at directly classifying clinically significant prostate cancer and approaches to quantify clinically relevant features such as segmenting and measuring the volume of the prostate gland (35).”

What the sources say

• Not checked yetcites [35] Li, 2022, Machine learning in prostate MRI for prostate cancer: current status and future opportunities, Diagnostics source

In the paper

“The platform was subsequently validated in an external cohort (61) and eventually cleared by the FDA (DEN200080, RRID:SCR_012945).”

What the sources say

• Not checked yetcites [61] da Silva, 2021, Independent real-world application of a clinical-grade automated prostate cancer detection system, J Pathol source
• Not checked yetcites DEN200080

In the paper

“Google developed LYmph Node Assistant (LYNA; ref. 46) and reported that pathologists who used the system were able to detect significantly more micrometastases with shorter review time.”

What the sources say

• Not checked yetcites [46] Steiner, 2018, Impact of deep learning assistance on the histopathologic review of lymph nodes for metastatic breast cancer, Am J Surg Pathol source

In the paper

“Treatment decisions in breast cancer frequently hinge on the detection of such lesions (44), but pathology review remains labor intensive and subject to significant interreader variability (arXiv:1703.02442).”

What the sources say

• Not checked yetcites [44] Giuliano, 2018, Eighth edition of the AJCC cancer staging manual: breast cancer, Ann Surg Oncol source
• Not checked yetcites arXiv:1703.02442

In the paper

“Sandbank and colleagues (50) have recently presented a deep learning model for classifying invasive and noninvasive breast cancer subtypes and generating 51 predicted clinical and morphologic features.”

What the sources say

• Not checked yetcites [50] Sandbank, 2022, Validation and real-world clinical application of an artificial intelligence algorithm for breast cancer detection in biopsies, NPJ Breast Cancer source

In the paper

“AI has been applied to each of these tasks, though these applications are generally further from clinical adoption with fewer FDA or EU clearances than the imaging-based detection applications described above.”

In the paper

“Ardila and colleagues (29) developed an approach to localize lung nodules and predict their likelihood of malignancy, where this model is now licensed by the AI company Aidence (30).”

What the sources say

• Not checked yetcites [29] Ardila, 2019, End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography, Nat Med source
• Not checked yetcites [30] 2022, Aidence and Google Health enter into collaboration to help improve lung cancer screening with AI

In the paper

“the Prostate Imaging-Cancer AI (PI-CAI) challenge enables participants to train on a multicenter, multivendor data set of 1,500 MRI prostate exams (36).”

What the sources say

• Not checked yetcites [36] Sunoqrot, 2022, Artificial intelligence for prostate MRI: open datasets, available applications, and grand challenges, Eur Radiol Exp source

In the paper

“Much of this work has now been integrated into Stratipath (59), a Swedishbased AI start-up focused on digital pathology.”

What the sources say

• Not checked yetcites [59] Stratipath Breast

In the paper

“Radiotherapy is a critical therapeutic approach for multiple cancer types, and an intense area of AI research, development, and commercialization.”

In the paper

“These efforts include histology (102) and CT-based (103) models for survival prediction, and the correlation of CT imaging features with histology risk factors (104).”

What the sources say

• Not checked yetcites [102] Yu, 2016, Predicting non-small cell lung cancer prognosis by fully automated microscopic pathology image features, Nat Commun source
• Not checked yetcites [103] Saad, 2023, Predicting benefit from immune checkpoint inhibitors in patients with non-small-cell lung cancer by CT-based ensemble deep learning: a retrospective study, Lancet Digit Health source
• Not checked yetcites [104] Nam, 2022, Histopathologic basis for a chest CT deep learning survival prediction model in patients with lung adenocarcinoma, Radiology source

In the paper

“This includes the use of AI at all steps of radiotherapy, including image segmentation, treatment planning, outcome prediction, quality assurance, and AIguided radiosurgery (116).”

What the sources say

• Not checked yetcites [116] Li, 2022, Artificial intelligence in radiotherapy, Semin Cancer Biol source

In the paper

“Additionally, CEmarked products exist for quantifying PD-L1 expression from IHC slides (109-111).”

What the sources say

• Not checked yetcites [109] Aiforia achieves CE-IVD mark for clinical PD-L1 AI model in lung cancer diagnostics
• Not checked yetcites [110] Mindpeak Solutions: plug-and-play AI solution to detect and quantify PD-L1 (SP263) within predefined ROIs in non-small-cell lung carcinoma (NSCLC)
• Not checked yetcites [111]

In the paper

“There have additionally been several validation studies of histopathology-based prognostic models (95,96), including a study demonstrating that a prognostic feature previously identified through AI analysis can be learned and used by pathologists (95).”

What the sources say

• Not checked yetcites [95] L'Imperio, 2023, Pathologist validation of a machine learning–derived feature for colon cancer risk stratification, JAMA Netw Open source
• Not checked yetcites [96] Skrede, 2020, Deep learning for prediction of colorectal cancer outcome: a discovery and validation study, Lancet Lond Engl source

In the paper

“Although there are some clinically available devices for these tasks, they tend to be further from clinical maturity than cancer detection and diagnosis tasks.”

In the paper

“Nagpal and colleagues (62) used a deep learning model to predict Gleason scores using H&E images from The Cancer Genome Atlas (TCGA) and reported significantly higher diagnostic accuracy compared with a cohort of 29 board-certified pathologists.”

What the sources say

• Not checked yetcites [62] Nagpal, 2019, Development and validation of a deep learning algorithm for improving Gleason scoring of prostate cancer, NPJ Digit Med source

In the paper

“P-NET (112) is a sparse deep learning model that processes molecular profiling data in a biologically informed, pathway-driven fashion to predict prostate disease states such as metastases, where the resulting prediction scores were shown to independently correlate with prognosis.”

What the sources say

• Not checked yetcites [112] Elmarakeby, 2021, Biologically informed deep neural network for prostate cancer discovery, Nature source

In the paper

“Significant efforts have also been made in using AI for histopathologic diagnosis in lung and colorectal cancers, though FDA-cleared devices for these tasks do not exist at the time of writing.”

In the paper

“The approach included weakly supervised training to predict response from H&E whole slide images, with subsequent application of interpretability approaches to identify features that correlate with higher response predictions, such as the presence of tumor-infiltrating lymphocytes (TILs) and necrosis.”

What the sources say

• Not checked yetcites [83] Ogier du Terrail, 2023, Federated learning for predicting histological response to neoadjuvant chemotherapy in triple-negative breast cancer, Nat Med source

In the paper

“Models such as BERT (80), GPT (81), and ChatGPT (arXiv:2303.08774)are rapidly diffusing into industry, and healthcare-specific models are also being developed (ref.82; arXiv:2106.03598,arXiv:1904.05342).”

What the sources say

• Not checked yetcites [80] Devlin, 2019, Pre-training of deep bidirectional transformers for language understanding
• Not checked yetcites [81] Brown, 2020, Language models are few-shot learners
• Not checked yetcites arXiv:2303.08774

In the paper

“Amgad and colleagues (84) explicitly enforce interpretability by using deep learning to extract human-interpretable features, and subsequently use these features to generate a risk score for breast cancer prognosis.”

What the sources say

• Not checked yetcites [84] Amgad, 2024, A population-level digital histologic biomarker for enhanced prognosis of invasive breast cancer, Nat Med source

In the paper

“Although the LLMs generated a wider range of treatment options compared with human experts, their recommendations often deviated from expert consensus.”

What the sources say

• Not checked yetcites [151] Benary, 2023, Leveraging large language models for decision support in personalized oncology, JAMA Netw Open source

In the paper

“CancerVar (131) is an AI platform that uses a deep learning framework to predict the oncogenicity of somatic variants using both functional and clinical features and may provide a robust alternative to manually curated databases, such as CiVIC (132) and OncoKB ( 133).”

What the sources say

• Not checked yetcites [131] Li, 2022, CancerVar: an artificial intelligence-empowered platform for clinical interpretation of somatic mutations in cancer, Sci Adv source
• Not checked yetcites [132] Griffith, 2017, CIViC is a community knowledgebase for expert crowdsourcing the clinical interpretation of variants in cancer, Nat Genet source
• Not checked yetcites [133] Chakravarty, 2017, OncoKB: a precision oncology knowledge base, JCO Precis Oncol

In the paper

“personalized cancer vaccines are being tested in a wide variety of clinical indications, and recent trial data indicate their potential role in combination with existing immune-checkpoint blockade (139).”

What the sources say

• Not checked yetcites [139] Khattak, 2023, Abstract CT001: a personalized cancer vaccine, mRNA-4157, combined with pembrolizumab versus pembrolizumab in patients with resected high-risk melanoma: efficacy and safety results from the randomized, open-label Phase 2 mRNA-4157-P201/Keynote-942 trial, Cancer Res source

In the paper

“Applied to 37 cancer types from the Pan-Cancer Analysis of Whole Genomes data set (128), Dig identified several potential cryptic splice sites, recurrent variants in noncoding regions, as well as mutations in rarely mutated genes.”

What the sources say

• Not checked yetcites [128] ICGC/TCGA Pan-Cancer Analysis of Whole Genomes Consortium, 2020, Pan-cancer analysis of whole genomes, Nature source

In the paper

“These challenges include obtaining and sharing robust medical data, risks to patient privacy, and ensuring fairness in AI performance across all patients.”

In the paper

“AI applications are also now advancing in the fields of image-guided radiotherapy, adaptive radiotherapy, stereotactic radiosurgery (119), and surgery in general, where autocontouring is similarly helpful (120).”

What the sources say

• Not checked yetcites [119] Kawamura, 2024, Revolutionizing radiation therapy: the role of AI in clinical practice, J Radiat Res (Tokyo) source
• Not checked yetcites [120] Mascagni, 2022, Computer vision in surgery: from potential to clinical value, NPJ Digit Med source

In the paper

“A 2022 meta-analysis reported on 10 published studies across multiple commercial and academic platforms and found that most applications offered comparable, if not superior performance to manual screening, with dramatic improvements in efficiency (149).”

What the sources say

• Not checked yetcites [149] Chow, 2023, Use of artificial intelligence for cancer clinical trial enrollment: a systematic review and meta-analysis, J Natl Cancer Inst source

In the paper

“A recent study by Radici and colleagues (117) evaluated the FDAcleared and CE-marked Limbus Contour (LC) AI platform for autocontouring within a single European tertiary cancer hospital and found that the technology significantly reduced the procedure time and interobserver variability.”

What the sources say

• Not checked yetcites [117] Radici, 2022, Implementation of a commercial deep learning-based auto segmentation software in radiotherapy: evaluation of effectiveness and impact on workflow, Life Basel Switz

In the paper

“only 6% of patients with cancer enroll in clinical trials (144) and there continue to be race-, sex-, and age-based disparities within clinical trials (145, 146).”

What the sources say

• Not checked yetcites [144] Unger, 2021, Nationally representative estimates of the participation of cancer patients in clinical research studies according to the commission on cancer, J Clin Oncol source
• Not checked yetcites [145] Murthy, 2004, Participation in cancer clini­cal trials: race-, sex-, and age-based disparities, JAMA source
• Not checked yetcites [146] Kwiatkowski, 2013, Inclusion of minorities and women in cancer clinical trials, a decade later: have we improved?, Cancer source

In the paper

“Transcriptomic-based algorithms have also been developed (136, 137), although these approaches will require expansion of clinical transcriptome profiling for assessments of generalizability.”

What the sources say

• Not checked yetcites [136] Grewal, 2019, Application of a neural network whole transcriptome-based pan-cancer method for diagnosis of primary and metastatic cancers, JAMA Netw Open source
• Not checked yetcites [137] Divate, 2022, Deep learning-based pan-cancer classification model reveals tissue-of-origin specific gene expression signatures, Cancers source

In the paper

“As more AI models reach the clinic, there are also challenges in effectively integrating AI applications into clinical workflows and assessing the ultimate impact on patients and healthcare systems.”

In the paper

“Despite these limitations, LLMs were able to correctly identify several important treatment strategies and even suggested some reasonable options that were not readily apparent to experts (151).”

What the sources say

• Not checked yetcites [151] Benary, 2023, Leveraging large language models for decision support in personalized oncology, JAMA Netw Open source

In the paper

“The method outperformed multiple other methods for driver identification in accuracy and power and ran multiple orders of magnitude faster.”

What the sources say

• Not checked yetcites [127] Sherman, 2022, Genome-wide mapping of somatic mutation rates uncovers drivers of cancer, Nat Biotechnol source

In the paper

“multiple groups have now reported on the use of LLMs to effectively extract structured data from clinical notes (150) and to match patients to clinical trials (arXiv:2303.16756,arXiv:2306.02077,arXiv:2304.07396, and arXiv:2308.02180).”

What the sources say

• Not checked yetcites [150] Agrawal, Large language models are few-shot clinical information extractors source
• Not checked yetcites arXiv:2303.16756
• Not checked yetcites arXiv:2306.02077

In the paper

“This is particularly important in cancer, where each site is limited in patient size, patient diversity, and its own set of embedded clinical practices.”

In the paper

“The interpretation of germline and somatic mutations observed in cancer continues to be a challenge, and new AI tools are also emerging to automatically annotate and interpret variants.”

In the paper

“Performance was generally promising, though there were cases in which the model arguably provided outdated information, such as investigational drug names (152).”

What the sources say

• Not checked yetcites [152] Schulte, 2023, Capacity of ChatGPT to identify guideline-based treatments for advanced solid tumors, Cureus

In the paper

“that risks driving providers away from clinical practice and impairing access to care.”

In the paper

“The clinical trial system also suffers from overall inefficiencies, where 20% of trials are prematurely terminated, many due to insufficient accrual of patients (147).”

What the sources say

• Not checked yetcites [147] Stensland, 2014, Adult cancer clinical trials that fail to complete: an epidemic?, J Natl Cancer Inst source

In the paper

“MD Anderson Cancer Center is also creating a comprehensive AI-based Radiation Planning Assistant for autocontouring and radiotherapy planning, and will primarily target low-and middleincome countries, where AI can provide scalable assistance (118).”

What the sources say

• Not checked yetcites [118] MD Anderson Cancer Center: Radiation Planning Assistant

In the paper

“This includes genomics-based algorithms, where targeted NGS algorithms have been rigorously evaluated and are currently being deployed and assessed for clinical utility (134, 135).”

What the sources say

• Not checked yetcites [134] Penson, 2020, Development of genome-derived tumor type prediction to inform clinical cancer care, JAMA Oncol source
• Not checked yetcites [135] Moon, 2023, Machine learning for genetics-based classification and treatment response prediction in cancer of unknown primary, Nat Med source

In the paper

“Beyond training, AI models must also be tested across multiple independent sites and evaluated for overall accuracy and potential bias (162).”

What the sources say

• Not checked yetcites [162] Karargyris, 2023, Federated benchmarking of medical artificial intelligence with MedPerf, Nat Mach Intell source

In the paper

“Rich molecular data are increasingly critical for personalized cancer treatment across cancer types.”

In the paper

“An additional long-standing diagnostic challenge is cell-oforigin prediction, which is particularly relevant for cancers of unknown primary where accurate origin information could significantly affect treatment strategy.”

In the paper

“Properly deployed, AI could increase efficiencies and patient diversity within the clinical trial system and assist physicians in matching patients to clinical trials.”

In the paper

“The third option is federated learning (FL), where data remain private at each institution, but machine learning models are jointly updated and shared in a distributed fashion (166, 167).”

What the sources say

• Not checked yetcites [166] Rieke, 2020, The future of digital health with federated learning, NPJ Digit Med source
• Not checked yetcites [167] Saldanha, 2022, Swarm learning for decentralized artificial intelligence in cancer histopathology, Nat Med source

In the paper

“These different data modalities are likely to contain distinct and complementary information; thus, multimodal AI is an active area of research in oncology (168, 169).”

What the sources say

• Not checked yetcites [168] Boehm, 2022, Harnessing multimodal data integration to advance precision oncology, Nat Rev Cancer source
• Not checked yetcites [169] Lipkova, 2022, Artificial intelligence for multimodal data integration in oncology, Cancer Cell source

In the paper

“Importantly, given this classification, randomized controlled clinical trials are not typically required by the FDA, though this can depend on the device's intended use.”

In the paper

“Neither the United States nor the EU currently has AI-specific regulatory approval pathways for AI-based medical tools.”

In the paper

“Although clinical trials in AI oncology remain rare, we are now starting to see an uptick in such trials.”

In the paper

“A particularly evolving and debated topic is the regulation of LLMs, and in which contexts and intended uses do they qualify as medical devices (191,192).”

What the sources say

• Not checked yetcites [191] Meskó, 2023, The imperative for regulatory oversight of large language models (or generative AI) in healthcare, NPJ Digit Med source
• Not checked yetcites [192] Gilbert, 2023, Large language model AI chatbots require approval as medical devices, Nat Med source

In the paper

“These efforts further emphasize the need for robust data sets, as the same data set and even model architecture can be used to support multiple clinical tasks.”

In the paper

“Rather, both regulate AI healthcare applications as medical devices and categorize such devices based on the level of risk posed to individual patients.”

In the paper

“Both of these considerations certainly apply in oncology, where potential sources of bias span the full patient care and AI life-cycle continuums.”

In the paper

“In clinical practice, there are known disparities in access to newer imaging technologies and patient outcomes for underserved populations, such as Black women (176-178).”

What the sources say

• Not checked yetcites [176] Christensen, 2023, Relationship between race and access to newer mammographic technology in women with medicare insurance, Radiology source
• Not checked yetcites [177] Rauscher, 2012, Disparities in screening mammography services by race/ethnicity and health insurance, J Womens Health source
• Not checked yetcites [178] Martini, 2022, Breast cancer disparities in outcomes; unmasking biological determinants associated with racial and genetic diversity, Clin Exp Metastasis source

In the paper

“An aspect that is closely tied to regulation is the evaluation of devices to ensure safe and effective performance, where clinical trials remain the gold standard in assessing new clinical interventions.”

In the paper

“The same study also found that breast cancer was the most commonly cited cancer type (31%), most commonly associated with radiology, providing further evidence of the advancing maturity of AI in mammography.”

What the sources say

• Not checked yetcites [194] Luchini, 2022, Artificial intelligence in oncology: current applications and future perspectives, Br J Cancer source

In the paper

“A critical component of AI evaluation and development more generally is ensuring transparency and reproducibility, allowing independent research groups to review and build on prior scientific progress (196).”

What the sources say

• Not checked yetcites [196] Haibe-Kains, 2020, Transparency and reproducibility in artificial intelligence, Nature source

In the paper

“The requirements for FDA clearance depend largely on the class and intended use of the device, with device classes ranging from I (lowest risk) to III (highest risk).”

In the paper

“Fairness and diversity have especially important considerations in such applications given that genetic ancestry may be causally related to some genetic traits and diseases (178, 183-186), as opposed to the social construct of race (173).”

What the sources say

• Not checked yetcites [178] Martini, 2022, Breast cancer disparities in outcomes; unmasking biological determinants associated with racial and genetic diversity, Clin Exp Metastasis source
• Not checked yetcites [183] Zhang, 2019, Characterization of frequently mutated cancer genes in Chinese breast tumors: a comparison of Chinese and TCGA cohorts, Ann Transl Med source
• Not checked yetcites [184] Mahal, 2020, Racial differences in genomic profiling of prostate cancer, N Engl J Med source
• Not checked yetcites [185] Gusev, 2016, Atlas of prostate cancer heritability in European and African-American men pinpoints tissue-specific regulation, Nat Commun source
• Not checked yetcites [186] Nassar, 2022, Ancestry-driven recalibration of tumor mutational burden and disparate clinical outcomes in response to immune checkpoint inhibitors, Cancer Cell source
• Not checked yetcites [173] Chen, 2023, Algorithmic fairness in artificial intelligence for medicine and healthcare, Nat Biomed Eng source

In the paper

“Although FL introduces a number of logistical and algorithmic complexities, progress has been made in narrowing the gap in model performance between FL and centralized learning (160), where FL has now been successfully applied to multiple cancer applications.”

What the sources say

• Not checked yetcites [160] Sheller, 2020, Federated learning in medicine: facilitating multi-institutional collaborations without sharing patient data, Sci Rep source

In the paper

“The vast majority of AI devices thus far, including those in oncology, fall under Class II, for which the FDA 510(k) pathway applies and requires the device to be”

What the sources say

• Not checked yetcites [187] Harvey, 2020, How the FDA regulates AI, Acad Radiol source

In the paper

“These algorithms are largely designed to assist clinicians in making diagnoses and/or quantify image-based properties (e.g., lesion size), but the clinician makes the final call.”

In the paper

“Altogether, there is growing awareness of the fairness of AI algorithms, including guidance from regulatory bodies, but there remains much room for improvement in reducing bias in AI and healthcare more broadly.”

In the paper

“The second option is deidentification and public release of data, usually via a consortium model.”

What the sources say

• Not checked yetcites [45] Ehteshami Bejnordi, 2017, Diagnostic assessment of deep learning algorithms for detection of lymph node metastases in women with breast cancer, JAMA source
• Not checked yetcites [165] Mullainathan, 2022, Solving medicine's data bottleneck: Nightingale Open Science, Nat Med source

In the paper

“For example, Luchini and colleagues (194) summarized 71 FDA-cleared devices in oncology (up to 2021) and found that radiology accounted for the majority of approved devices (54.9%), followed by pathology (19.7%).”

What the sources say

• Not checked yetcites [194] Luchini, 2022, Artificial intelligence in oncology: current applications and future perspectives, Br J Cancer source

In the paper

“This work leverages the TCGA database, which has become a critical multimodal data set to advance AI efforts.”

In the paper

“Such”

What the sources say

• Not checked yetcites arXiv:2108.07258

In the paper

“Within the EU, there is no one central agency that performs device assessments; rather, assessments are handled by accredited organizations, which are empowered to issue CE marks.”

In the paper

“Conversely, there are efforts to generate more diverse public data sets, such as EMBED (180), and instances where AI performance has generalized well across different patient populations (9,10,12).”

What the sources say

• Not checked yetcites [180] Jeong, 2023, The EMory BrEast imaging Dataset (EMBED): a racially diverse, granular dataset of 3.4 million screening and diagnostic mammographic images, Radiol Artif Intell source
• Not checked yetcites [9] McKinney, 2020, International evaluation of an AI system for breast cancer screening, Nature source
• Not checked yetcites [10] Lotter, 2021, Robust breast cancer detection in mammography and digital breast tomosynthesis using an annotation-efficient deep learning approach, Nat Med source

In the paper

“Most of the described AI applications fall under the purview of regulatory agencies, depending on regional regulations and the intended use of the device.”

In the paper

“The precise form of the AI results requires careful consideration of the clinical end goal and the intended use of the device.”

In the paper

“This is especially important for assessing potential biases in data sets and generalizability across different clinical settings.”

In the paper

“Critically, although AI will ideally improve healthcare outcomes, its biggest impact may ultimately be in improving resource allocation and access to quality care.”

In the paper

“Second, the healthcare value of AI deployment needs to be rigorously assessed and combined with aligned payment structures.”

In the paper

“Such workflows require integration into clinical systems, which can be especially challenging for proprietary systems with limited customization.”

In the paper

“This will require a greater investment in real-world studies and ideally randomized clinical trials, with an emphasis on critically evaluating AI generalizability across patient populations and clinical sites.”

In the paper

“Nonetheless, the future of AI oncology will not be driven by technology innovation alone.”

In the paper

“Rather, we see several fundamental requirements that will drive the future of AI in oncology (Fig. 4).”

In the paper

“First, there is a strong need for an objective assessment of AI's impact on patient outcomes, defined in cancer-specific, clinically meaningful metrics.”

In the paper

“Finally, stakeholders must adopt objective processes for continuously monitoring AI impact and measuring patient safety.”

In the paper

“Realizing these benefits may require structural changes and new clinical workflows.”

In the paper

“Arguably, the greatest strength of data-driven, deep learning approaches is the ability to continually improve over time.”

In the paper

“Healthcare value is defined as the health outcome achieved per dollar spent (214) and is a critical metric for making informed healthcare investments.”

What the sources say

• Not checked yetcites [214] Porter, 2010, What is value in health care?, N Engl J Med source

In the paper

“However, realizing these benefits requires well-defined regulatory processes and joint efforts to continuously monitor AI devices over the long-term.”

In the paper

“Beyond initial hardware considerations, data pipelines must also be developed to efficiently route data to AI systems and subsequently make the AI results available to clinicians.”

In the paper

“Common approaches to XAI involve post hoc explanations, such as generating saliency maps that highlight aspects of the input that most contribute to the model's prediction (204).”

What the sources say

• Not checked yetcites [204] van der Velden, 2022, Explainable artificial intelligence (XAI) in deep learning-based medical image analysis, Med Image Anal source

In the paper

“The field of AI in oncology is developing at a rapid pace.”

In the paper

“Even with effective infrastructure and AI outputs, monetization and liability questions can hinder widespread adoption, where the vast majority of commercially available AI devices are not currently reimbursed (210).”

What the sources say

• Not checked yetcites [210] Parikh, 2022, Paying for artificial intelligence in medicine, NPJ Digit Med source

In the paper

“Additional guidelines include SPIRIT-AI (Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence), which provides an extended set of 15 items for AI-focused protocol documents (199), and CONSORT-AI (Consolidated Standards of Reporting of Trials-Artificial Intelligence), which provides an extended set of 14 items for reporting on AI-focused clinical trials (200).”

What the sources say

• Not checked yetcites [199] Rivera, 2020, Guidelines for clinical trial protocols for interventions involving artificial intelligence: the SPIRIT-AI extension, Lancet Digit Health source
• Not checked yetcites [200] Liu, 2020, Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension, Nat Med source

In the paper

“This includes MINIMAR (MINimum Information for Medical AI Reporting), which defines a minimal set of reporting elements covering study population and setting, patient demographics, model architecture, and model evaluation (198)”

What the sources say

• Not checked yetcites [198] Hernandez-Boussard, 2020, MINimum Information for Medical AI Reporting): developing reporting standards for artificial intelligence in health care, J Am Med Inform Assoc source

In the paper

“Finally, there is an increasing use of AI in analyzing real-world data (RWD), and the European Society for Medical Oncology Guidance for Reporting Oncology Real-World Evidence (ESMO-GROW) now provides publication guidelines regarding the use of RWD in oncology, along with specific AIfocused reporting guidelines (201).”

What the sources say

• Not checked yetcites [201] Castelo-Branco, 2023, ESMO guidance for reporting oncology real-world evidence (GROW), Ann Oncol source

In the paper

“This necessity applies even to”

In the paper

“Realizing the potential of AI in oncology requires effective integration into clinical workflows, which presents both logistical and scientific challenges (202).”

What the sources say

• Not checked yetcites [202] Chua, 2021, Artificial intelligence in oncology: path to implementation, Cancer Med source

In the paper

“As such, explainable AI (XAI) is a highly active area of research focused on conveying how an AI model arrived at its prediction.”

In the paper

“AI applications require significant computational infrastructure (either on-premise or cloud-based) and highly skilled engineers-both of which are likely to extend already stretched healthcare information technology (IT) budgets.”