Artificial intelligence in pancreatic cancer

Huang, Bowen; Huang, Haoran; Zhang, Shuting; Zhang, Dingyue; Shi, Qingya; Liu, Jianzhou; Guo, Junchao

doi:10.7150/thno.77949

Cited by 36 publications

(32 citation statements)

References 277 publications

(394 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This could be particularly valuable for the treatment of challenging diseases such PC. Despite still in its infancy, AI applications to PC have demonstrated promising results from diagnosis to treatment (28)(29)(30). Based on these preliminary data, we do believe that such a new technology may give a relevant support to clinicians especially in the context of a multidisciplinary approach, where all specialized medical figures may be supported by AI technology in order to guarantee the most adequate diagnostic-therapeutic pathway.…”

Section: Discussionmentioning

confidence: 99%

The role of the multidisciplinary tumor board (MDTB) in the assessment of pancreatic cancer diagnosis and resectability: A tertiary referral center experience

Quero

Sio²,

Fiorillo³

et al. 2023

Front. Surg.

View full text Add to dashboard Cite

BackgroundThe introduction of multidisciplinary tumor boards (MDTBs) for the diagnostic and therapeutic pathway of several oncological disease significantly ameliorated patients' outcomes. However, only few evidences are currently present on the potential impact of the MDTB on pancreatic cancer (PC) management. Aim of this study is to report how MDTB may influence PC diagnosis and treatment, with particular focus on PC resectability assessment and the correspondence between MDTB definition of resectability and intraoperative findings.MethodsAll patients with a proven or suspected diagnosis of PC discussed at the MDTB between 2018 and 2020 were included in the study. An evaluation of diagnosis, tumor response to oncological/radiation therapy and resectability before and after the MDTB was conducted. Moreover, a comparison between the MDTB resectability assessment and the intraoperative findings was performed.ResultsA total of 487 cases were included in the analysis: 228 (46.8%) for diagnosis evaluation, 75 (15.4%) for tumor response assessment after/during medical treatment, 184 (37.8%) for PC resectability assessment. As a whole, MDTB led to a change in treatment management in 89 cases (18.3%): 31/228 (13.6%) in the diagnosis group, 13/75 (17.3%) in the assessment of treatment response cohort and 45/184 (24.4%) in the PC resectability evaluation group. As a whole, 129 patients were given indication to surgery. Surgical resection was accomplished in 121 patients (93.7%), with a concordance rate of resectability between MDTB discussion and intraoperative findings of 91.5%. Concordance rate was 99% for resectable lesions and 64.3% for borderline PCs.ConclusionsMDTB discussion consistently influences PC management, with significant variations in terms of diagnosis, tumor response assessment and resectability. In this last regard, MDTB discussion plays a key role, as demonstrated by the high concordance rate between MDTB resectability definition and intraoperative findings.

show abstract

Section: Discussionmentioning

confidence: 99%

The role of the multidisciplinary tumor board (MDTB) in the assessment of pancreatic cancer diagnosis and resectability: A tertiary referral center experience

Quero

Sio²,

Fiorillo³

et al. 2023

Front. Surg.

View full text Add to dashboard Cite

show abstract

“…At the end, this may be a step towards improved personalized oncology therapy concepts, which will in the future integrate large clinical, radiological and pathological data sets using artificial intelligence. 43…”

Section: Discussionmentioning

confidence: 99%

“…Furthermore, we provide an exceptional annotated liver dataset for the development and validation of deep learning algorithms which we provided to the scientific community. At the end, this may be a step towards improved personalized oncology therapy concepts, which will in the future integrate large clinical, radiological and pathological data sets using artificial intelligence 43 …”

Section: Discussionmentioning

confidence: 99%

Implementation of deep learning in liver pathology optimizes diagnosis of benign lesions and adenocarcinoma metastasis

Kriegsmann

Steinbuß

et al. 2023

Clinical & Translational Med

View full text Add to dashboard Cite

Introduction Differentiation of histologically similar structures in the liver, including anatomical structures, benign bile duct lesions, or common types of liver metastases, can be challenging with conventional histological tissue sections alone. Accurate histopathological classification is paramount for the diagnosis and adequate treatment of the disease. Deep learning algorithms have been proposed for objective and consistent assessment of digital histopathological images. Materials and methods In the present study, we trained and evaluated deep learning algorithms based on the EfficientNetV2 and ResNetRS architectures to discriminate between different histopathological classes. For the required dataset, specialized surgical pathologists annotated seven different histological classes, including different non‐neoplastic anatomical structures, benign bile duct lesions, and liver metastases from colorectal and pancreatic adenocarcinoma in a large patient cohort. Annotation resulted in a total of 204.159 image patches, followed by discrimination analysis using our deep learning models. Model performance was evaluated on validation and test data using confusion matrices. Results Evaluation of the test set based on tiles and cases revealed overall highly satisfactory prediction capability of our algorithm for the different histological classes, resulting in a tile accuracy of 89% (38 413/43 059) and case accuracy of 94% (198/211). Importantly, the separation of metastasis versus benign lesions was certainly confident on case level, confirming the classification model performed with high diagnostic accuracy. Moreover, the whole curated raw data set is made publically available. Conclusions Deep learning is a promising approach in surgical liver pathology supporting decision making in personalized medicine.

show abstract

“…The lack of biomarkers for early detection of pancreatic ductal adenocarcinoma (PDAC) represents a hot topic of research, and their assessment in clinical practice could represent a turning point in the fight against this lethal disease (Jacks, et al 2022). To date, high costs for screening programs and low reliability of available specific PDAC markers still give a large advantage to the disease that continues to have a poor prognosis as it is too often identified in its advanced stage (Huang et al 2022). According to what the World Health Organization reported, to be used in clinical practice a marker should satisfy the REASSURED criteria (Land et al 2019).…”

Section: Introductionmentioning

confidence: 99%

Coupling magnetic levitation of graphene oxide–protein complexes with blood levels of glucose for early detection of pancreatic adenocarcinoma

et al. 2023

View full text Add to dashboard Cite

Introduction Pancreatic adenocarcinoma (PDAC) has a poor prognosis since often diagnosed too late. Dyslipidemia and hyperglycemia are considered risk factors, but the presence of the tumor itself can determine the onset of these disorders. Therefore, it is not easy to predict which subjects with diabetes or dyslipidemia will develop or have already developed PDAC. Over the past decade, tests based on the use of nanotechnology, alone or coupled with common laboratory tests (e.g., hemoglobin levels), have proven useful in aiding the diagnosis of PDAC. Tests based on magnetic levitation (MagLev) have demonstrated high diagnostic accuracy in compliance with the REASSURED criteria. Here, we aimed to assess the ability of the MagLev test in detecting PDAC when coupled with the blood levels of glycemia, cholesterol, and triglycerides. Methods Blood samples from 24 PDAC patients and 22 healthy controls were collected. Human plasma was let to interact with graphene oxide (GO) nanosheets and the emerging coronated systems were put in the MagLev device. Outcomes from Maglev experiments were coupled to glycemia, cholesterol, and triglycerides levels. Linear discriminant analysis (LDA) was carried out to evaluate the classification ability of the test in terms of specificity, sensitivity, and global accuracy. Statistical analysis was performed with Matlab (MathWorks, Natick, MA, USA, Version R2022a) software. Results The positions of the levitating bands were measured at the starting point (i.e., as soon as the cuvette containing the sample was subjected to the magnetic field). Significant variations in the starting position of levitating nanosystems in controls and PDACs were detected. The combination of the MagLev outcomes with the blood glycemic levels returned the best value of global accuracy (91%) if compared to the coupling with those of cholesterol and triglycerides (global accuracy of ~ 77% and 84%, respectively). Conclusion If confirmed by further studies on larger cohorts, a multiplexed Maglev-based nanotechnology-enabled blood test could be employed as a screening tool for PDAC in populations with hyperglycemia.

show abstract

Artificial intelligence in pancreatic cancer

Cited by 36 publications

References 277 publications

The role of the multidisciplinary tumor board (MDTB) in the assessment of pancreatic cancer diagnosis and resectability: A tertiary referral center experience

The role of the multidisciplinary tumor board (MDTB) in the assessment of pancreatic cancer diagnosis and resectability: A tertiary referral center experience

Implementation of deep learning in liver pathology optimizes diagnosis of benign lesions and adenocarcinoma metastasis

Coupling magnetic levitation of graphene oxide–protein complexes with blood levels of glucose for early detection of pancreatic adenocarcinoma

Contact Info

Product

Resources

About