Recent advances and applications of artificial intelligence in 3D bioprinting

Biophysics Reviews, Journal Year: 2024, Volume and Issue: 5(3)

Published: July 19, 2024

3D bioprinting techniques enable the precise deposition of living cells, biomaterials, and biomolecules, emerging as a promising approach for engineering functional tissues organs. Meanwhile, recent advances in researchers to build vitro models with finely controlled complex micro-architecture drug screening disease modeling. Recently, artificial intelligence (AI) has been applied different stages bioprinting, including medical image reconstruction, bioink selection, printing process, both classical AI machine learning approaches. The ability handle datasets, make computations, learn from past experiences, optimize processes dynamically makes it an invaluable tool advancing bioprinting. review highlights current integration discusses future approaches harness synergistic capabilities developing personalized

Language: Английский

Advances and Challenges in 3D Bioprinted Cancer Models: Opportunities for Personalized Medicine and Tissue Engineering

Polymers, Journal Year: 2025, Volume and Issue: 17(7), P. 948 - 948

Published: March 31, 2025

Cancer is the second leading cause of death worldwide, after cardiovascular disease, claiming not only a staggering number lives but also causing considerable health and economic devastation, particularly in less-developed countries. Therapeutic interventions are impeded by differences patient-to-patient responses to anti-cancer drugs. A personalized medicine approach crucial for treating specific patient groups includes using molecular genetic screens find appropriate stratifications patients who will respond (and those not) treatment regimens. However, information on which risk stratification method can be used hone cancer types likely responders agent remains elusive most cancers. Novel developments 3D bioprinting technology have been widely applied recreate relevant bioengineered tumor organotypic structures capable mimicking human tissue microenvironment or adequate drug high-throughput screening settings. Parts autogenously printed form tissues computer-aided design concept where multiple layers include different cell compatible biomaterials build configurations. Patient-derived stromal cells, together with material, extracellular matrix proteins, growth factors, create bioprinted models that provide possible platform new therapies advance. Both natural synthetic biopolymers encourage cells biological materials models/implants. These may facilitate physiologically cell-cell cell-matrix interactions heterogeneity resembling real tumors.

Language: Английский

Leveraging advanced graph neural networks for the enhanced classification of post anesthesia states to aid surgical procedures

PLoS ONE, Journal Year: 2025, Volume and Issue: 20(4), P. e0320299 - e0320299

Published: April 25, 2025

Anesthesia plays a pivotal role in modern surgery by facilitating controlled states of unconsciousness. Precise control is crucial for safe and pain-free surgeries. Monitoring anesthesia depth accurately essential to guide anesthesiologists, optimize drug usage, mitigate postoperative complications. This study focuses on enhancing the classification performance anesthesia-induced transitions between wakefulness deep sleep into eight classes leveraging advanced graph neural network (GNN). The research combines seven datasets single dataset comprising 290 samples investigates key brain regions, develop robust framework. Initially, augmented using Synthetic Minority Over-sampling Technique (SMOTE) expand sample size 1197. A graph-based approach employed get intricate relationships features, constructing with 1197 nodes 714,610 edges, where represent data edges are connections nodes. connection (edge weight) calculated Spearman correlation coefficient matrix. An optimized GNN model developed through an ablation hyperparameters, achieving accuracy 92.8%. model’s further evaluated against one-dimensional (1D) CNN, six machine learning models, demonstrating superior capabilities small imbalanced datasets. Additionally, we proposed different datasets, observing no decline performance. work advances understanding states, providing valuable tool improved management.

Language: Английский