The Application of the Preoperative Image-Guided 3D Visualization Supported by Machine Learning to the Prediction of Organs Reconstruction During Pancreaticoduodenectomy via a Head-Mounted Displays DOI Creative Commons
Klaudia Proniewska,

Radek Kolecki,

Aneta Grochowska

et al.

Lecture notes in computer science, Journal Year: 2023, Volume and Issue: unknown, P. 321 - 344

Published: Jan. 1, 2023

Early pancreatic cancer diagnosis and therapy drastically increase the chances of survival. Tumor visualization using CT scan images is an important part these processes. In this paper, we apply Mixed Reality (MR) Artificial Intelligence, in particular, Machine Learning (ML) to prepare image-guided 3D models a population oncology patients. Object detection was based on convolution neural network, i.e. You Only Look Once (YOLO) version 7 algorithm, while semantic segmentation has been done with 3D-UNET algorithm. Next, holographic model as interactive, MR object performed Microsoft HoloLens2. The results indicated that proposed ML-based approach can precisely segment pancreas along suspected lesions, thus providing reliable tool for diagnostics surgical planning, especially when considering organ reconstruction during pancreaticoduodenectomy.

Language: Английский

Cardiac Healthcare Digital Twins Supported by Artificial Intelligence-Based Algorithms and Extended Reality—A Systematic Review DOI Open Access

Zofia Rudnicka,

Klaudia Proniewska, Mark D. Perkins

et al.

Electronics, Journal Year: 2024, Volume and Issue: 13(5), P. 866 - 866

Published: Feb. 23, 2024

Recently, significant efforts have been made to create Health Digital Twins (HDTs), for clinical applications. Heart modeling is one of the fastest-growing fields, which favors effective application HDTs. The HDTs will be increasingly widespread in future healthcare services and has huge potential form part mainstream medicine. However, it requires development both models algorithms analysis medical data, advances Artificial Intelligence (AI)-based already revolutionized image segmentation processes. Precise lesions may contribute an efficient diagnostics process a more selection targeted therapy. In this systematic review, brief overview recent achievements HDT technologies field cardiology, including interventional was conducted. were studied taking into account Extended Reality (XR) AI, as well data security, technical risks, ethics-related issues. Special emphasis put on automatic study, 253 literature sources taken account. It appears that improvements processing focus imaging addition three-dimensional (3D) pictures reconstruct anatomy heart torso can displayed XR-based devices. This diagnostics. combination XR, HDT-based solution help avoid errors serve universal methodology personalized cardiology. Additionally, we describe applications, limitations, further research directions.

Language: Английский

Citations

15
Therapeutic Approaches in Pancreatic Cancer: Recent Updates DOI Creative Commons
Lokender Kumar, Sanjay Kumar, Sandeep Kumar

et al.

Biomedicines, Journal Year: 2023, Volume and Issue: 11(6), P. 1611 - 1611

Published: June 1, 2023

Cancer is a significant challenge for effective treatment due to its complex mechanism, different progressing stages, and lack of adequate procedures screening identification. Pancreatic cancer typically identified in advanced progression phase with low survival ~5 years. Among cancers, pancreatic also considered high mortality-causing casualty over other accidental or disease-based mortality, it ranked seventh among all mortality-associated cancers globally. Henceforth, developing diagnostic early detection, understanding cancer-linked mechanisms, various therapeutic strategies are crucial. This review describes the recent development progression, approaches, including molecular techniques biomedicines effectively treating cancer.

Language: Английский

Citations

16
Lésions précancéreuses du pancréas, chirurgie prophylactique pancréatique DOI

G. Désolneux,

Fanny Castanet

Bulletin du Cancer, Journal Year: 2025, Volume and Issue: 112(3), P. 263 - 269

Published: March 1, 2025

Language: Английский

Citations

0
Advancing Precision Medicine: VAE Enhanced Predictions of Pancreatic Cancer Patient Survival in Local Hospital DOI Creative Commons
Yuan Wang,

Chenbi Li,

Zeheng Wang

et al.

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 3428 - 3436

Published: Jan. 1, 2024

In this research, we address the urgent need for accurate prediction of in-hospital survival periods patients diagnosed with pancreatic cancer (PC), a disease notorious its late-stage diagnosis and dismal rates.Utilizing machine learning (ML) technologies, focus on application Variational Autoencoders (VAE) data augmentation ensemble techniques enhancing predictive accuracy.Our dataset comprises biochemical blood test (BBT) results from stage II/III PC patients, which is limited in size, making VAE's capability particularly valuable.The study employs several ML models, including Elastic Net (EN), Decision Trees (DT), Radial Basis Function Support Vector Machine (RBF-SVM), evaluates their performance using metrics such as Mean Absolute Error (MAE) Squared (MSE).Our findings reveal that EN, DT, RBF-SVM are most effective models within VAE-augmented framework, showing substantial improvements accuracy.An approach further optimized results, reducing MAE to approximately 10 days.These advancements hold significant implications field precision medicine, enabling more targeted therapeutic interventions optimizing healthcare resource allocation.The can also serve foundational step towards personalized solutions patients.

Language: Английский

Citations

2
Health Digital Twins Supported by Artificial Intelligence-based Algorithms and Extended Reality in Cardiology DOI Creative Commons

Zofia Rudnicka,

Klaudia Proniewska, Mark D. Perkins

et al.

arXiv (Cornell University), Journal Year: 2024, Volume and Issue: unknown

Published: Jan. 1, 2024

Recently, significant efforts have been made to create Health Digital Twins (HDTs), digital twins for clinical applications. Heart modeling is one of the fastest-growing fields, which favors effective application HDTs. The HDTs will be increasingly widespread in future healthcare services and has a huge potential form part mainstream medicine. However, it requires development both models algorithms analysis medical data, advances Artificial Intelligence (AI) based already revolutionized image segmentation processes. Precise lesions may contribute an efficient diagnostics process more selection targeted therapy. In this paper, brief overview recent achievements HDT technologies field cardiology, including interventional cardiology was conducted. were studied taking into account Extended Reality (XR) AI, as well data security, technical risks, ethics-related issues. Special emphasis put on automatic It appears that improvements processing focus imaging addition three-dimensional (3D) pictures reconstruct anatomy heart torso can displayed XR-based devices. This diagnostics. combination XR, HDT-based solution help avoid errors serve universal methodology personalized cardiology. Additionally, we describe applications, limitations, further research directions.

Language: Английский

Citations

1
Risk Assessment and Radiomics Analysis in Magnetic Resonance Imaging of Pancreatic Intraductal Papillary Mucinous Neoplasms (IPMN) DOI Creative Commons

Federica Flammia,

Roberta Fusco,

Sonia Triggiani

et al.

Cancer Control, Journal Year: 2024, Volume and Issue: 31

Published: Jan. 1, 2024

Intraductal papillary mucinous neoplasms (IPMNs) are a very common incidental finding during patient radiological assessment. These lesions may progress from low-grade dysplasia (LGD) to high-grade (HGD) and even pancreatic cancer. The IPMN progression risk grows with time, so discontinuation of surveillance is not recommended. It important identify imaging features that suggest LGD IPMNs, thus, distinguish only require careful those need surgical resection. know the management guidelines especially indications for surgery, be able point out in report findings malignant degeneration. tools employed diagnosis assessment Computed Tomography (CT) Magnetic Resonance Imaging (MRI) contrast medium. According latest European guidelines, MRI method choice follow-up patients since this tool has highest sensitivity detecting mural nodules intra-cystic septa. plays key role worrisome high-risk stigmata, which associated IPMNs Nowadays, main limit diagnostic ability precursor In context, increasing attention being given artificial intelligence (AI) radiomics analysis. However, these remain an exploratory phase, considering limitations currently published studies. Key limits include noncompliance AI best practices, workflow standardization, clear reporting study methodology, including segmentation data balancing. it useful note type as morphological features, size, rate growth, wall, septa nodules, on surgery based. should reported time suggested according guidelines.

Language: Английский

Citations

1
Pancreatic Ductal Adenocarcinoma: Update of CT-Based Radiomics Applications in the Pre-Surgical Prediction of the Risk of Post-Operative Fistula, Resectability Status and Prognosis DOI Open Access

Giulia Pacella,

Maria Chiara Brunese,

Eleonora D’Imperio

et al.

Journal of Clinical Medicine, Journal Year: 2023, Volume and Issue: 12(23), P. 7380 - 7380

Published: Nov. 28, 2023

Pancreatic ductal adenocarcinoma (PDAC) is the seventh leading cause of cancer-related deaths worldwide. Surgical resection main driver to improving survival in resectable tumors, while neoadjuvant treatment based on chemotherapy (and radiotherapy) best option-treatment for a non-primally disease. CT-based imaging has central role detecting, staging, and managing PDAC. As several authors have proposed radiomics risk stratification patients undergoing surgery PADC, this narrative review, we explored actual fields interest tools PDAC built pre-surgical clinical variables, obtain more objective reliable predictors. The PubMed database was searched papers published English language no earlier than January 2018. We found 301 studies, 11 satisfied our research criteria. Of those included, four were resectability status prediction, three preoperative pancreatic fistula (POPF) prediction. Most studies retrospective. It possible conclude that many performing models been developed get predictive information evaluation. However, all retrospective, lacking further external validation prospective multicentric cohorts. Furthermore, expression results should be standardized automatized applicable practice.

Language: Английский

Citations

3
Prognostic Assessment of Gastropancreatic Neuroendocrine Neoplasm: Prospects and limits of Radiomics DOI Creative Commons
Federica De Muzio, Fabio Pellegrino, Roberta Fusco

et al.

Diagnostics, Journal Year: 2023, Volume and Issue: 13(18), P. 2877 - 2877

Published: Sept. 7, 2023

Neuroendocrine neoplasms (NENs) are a group of lesions originating from cells the diffuse neuroendocrine system. NENs may involve different sites, including gastrointestinal tract (GEP-NENs). The incidence and prevalence GEP-NENs has been constantly rising thanks to increased diagnostic power imaging immuno–histochemistry. Despite plethora biochemical markers techniques, prognosis therapeutic choice in still represents challenge, mainly due great heterogeneity terms tumor clinical behavior. concept that biomedical images contain information about tissue pathological processes invisible human eye is now well established. From this substrate comes idea radiomics. Computational analysis achieved promising results several oncological settings, use radiomics types growing field research, yet with conflicting results. aim narrative review provide comprehensive update on role GEP-NEN management, focusing main aspects analyzed by most existing reports: predicting grade, distinguishing NET other tumors, assessment.

Language: Английский

Citations

1
Future Perspectives on Radiomics in Acute Liver Injury and Liver Trauma DOI Open Access

Maria Chiara Brunese,

Pasquale Avella, Micaela Cappuccio

et al.

Journal of Personalized Medicine, Journal Year: 2024, Volume and Issue: 14(6), P. 572 - 572

Published: May 27, 2024

Background: Acute liver injury occurs most frequently due to trauma, but it can also occur because of sepsis or drug-induced injury. This review aims analyze artificial intelligence (AI)’s ability detect and quantify injured areas in adults pediatric patients. Methods: A literature analysis was performed on the PubMed Dataset. We selected original articles published from 2018 2023 cohorts with ≥10 Results: Six studies counting 564 patients were collected, including 170 (30%) children 394 adults. Four (66%) reported AI application after one (17%) sepsis, chemotherapy. In five (83%) studies, Computed Tomography performed, while (17%), FAST-UltraSound performed. The a high diagnostic performance; particular, three specificity rate > 80%. Conclusions: Radiomics models seem reliable applicable clinical practice affected by acute Further are required achieve larger validation cohorts.

Language: Английский

Citations

0
Diagnostic Value of Contrast-Enhanced Dual-Energy Computed Tomography in the Pancreatic Parenchymal and Delayed Phases for Pancreatic Cancer DOI Creative Commons
Yusuke Kurita, Daisuke Utsunomiya, Kensuke Kubota

et al.

Tomography, Journal Year: 2024, Volume and Issue: 10(10), P. 1591 - 1604

Published: Oct. 7, 2024

: The usefulness of dual-energy computed tomography (DECT) for low absorption in the parenchymal phase and contrast effects delayed pancreatic cancer is not clear. Therefore, diagnostic capability low-KeV images obtained using DECT phases was evaluated quantitatively qualitatively.

Language: Английский

Citations

0