Submit to Life Review for Life Propose a Special Issue

Journal Menu

Journal Browser

► Journal Browser

Current and Future Perspectives of Artificial Intelligence in Medicine

Print Special Issue Flyer
Special Issue Editors
Special Issue Information
Keywords
Benefits of Publishing in a Special Issue
Published Papers

A special issue of Life (ISSN 2075-1729). This special issue belongs to the section "Medical Research".

Deadline for manuscript submissions: closed (21 November 2025) | Viewed by 17229

Share This Special Issue

Special Issue Editors

Dr. Giuseppe Consorti

E-Mail Website
Guest Editor

Department of Maxillofacial Surgery, Azienda Ospedaliera Universitaria, Ospedali Riuniti di Ancona, Umberto, Italy
Interests: orthognathic surgery; facial aesthetics surgery; head and neck surgery; craniofacial surgery; cleft lip and palate; orbit; facial trauma and surgery; facial fracture; cancer; stem cell; oncology; reconstructive surgery; microsurgery; tumor resection; squamous cell carcinoma; malformation; rare disease; craniofacial malformation; new technologies; CAD-CAM; genetics; oral surgery
Special Issues, Collections and Topics in MDPI journals

Dr. Lisa Catarzi

E-Mail Website
Guest Editor

1. Maxillofacial Surgery, Careggi University Hospital, 50139 Florence, Italy
2. Department of Medical Biotechnology, University of Siena, 53100 Siena, Italy
Interests: maxillofacial surgery; maxillofacial trauma; orbital reconstruction; microsurgery; head and neck reconstruction; reconstructive surgery
Special Issues, Collections and Topics in MDPI journals

Dr. Guido Gabriele

E-Mail Website
Guest Editor

Maxillofacial Surgery Unit, Department of Medical Biotechnologies, University of Siena, 53100 Siena, Italy
Interests: maxillofacial surgery; head and neck surgery; microsurgery; lymphedema; supramicorsurgery
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Artificial intelligence (AI) holds the potential to revolutionize healthcare across all medical domains, marking a pivotal moment for the field of medicine. Traditionally, medical specialties have relied heavily on human interaction and innovation, but physicians are now adapting to incorporate AI as a valuable tool in patient care. AI brings the promise of ensuring patient safety, enhancing autonomy, and providing timely medical assistance, especially in remote areas, while simultaneously reducing the administrative workload, screen time, and the risk of professional burnout for healthcare providers. Moreover, AI has the capability of decreasing medical errors and enhancing diagnostic precision by leveraging algorithms and software to integrate, analyze, and interpret vast amounts of medical data. By automating repetitive tasks, AI can afford healthcare personnel more time to focus on building stronger doctor–patient relationships, emphasizing personalized care, communication, empathy, and trust during times of illness—essential aspects of care that cannot be replaced by AI. However, there remains a need to standardize research in this field to ensure the quality of scientific evidence, understand its benefits and drawbacks, and expedite its integration into mainstream medical practices.

Dr. Giuseppe Consorti
Dr. Lisa Catarzi
Dr. Guido Gabriele
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Life is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

artificial intelligence
machine learning
deep learning
medical practices
artificial neural network
computational health informatics
predictive analytics in healthcare
ethics in AI for medical applications

Benefits of Publishing in a Special Issue

Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
Reprint: MDPI Books provides the opportunity to republish successful Special Issues in book format, both online and in print.

Further information on MDPI's Special Issue policies can be found here.

Published Papers (9 papers)

Download All Papers

Order results

Result details

Show export options Show export options

Select all

Export citation of selected articles as:

Research

Jump to: Review, Other

16 pages, 5966 KB

Open AccessArticle

Low-Dose CT Quality Assurance at Scale: Automated Detection of Overscanning, Underscanning, and Image Noise

by Patrick Wienholt, Alexander Hermans, Robert Siepmann, Christiane Kuhl, Daniel Pinto dos Santos, Sven Nebelung and Daniel Truhn

Life 2026, 16(1), 152; https://doi.org/10.3390/life16010152 - 16 Jan 2026

Viewed by 529

Abstract

Automated quality assurance is essential for low-dose computed tomography (LDCT) lung screening, yet manual checks strain clinical workflows. We present a fully automated artificial intelligence tool that quantifies scan coverage and image noise in LDCT without user input. Lungs and the aorta are segmented to measure cranial/caudal over- and underscanning, and noise is computed as the standard deviation of Hounsfield units (HUs) within descending aortic blood, normalized to a 1

{mm}^{3}

voxel. Performance was verified in a reader study of 98 LDCT scans from the National Lung Screening Trial (NLST), and then applied to 38,834 NLST scans reconstructed with a standard kernel. In the reader study, lung masks were rated ≥“Nearly Perfect” in 90.8% and aorta-blood masks in 96.9% of cases. Across 38,834 scans, mean overscanning distances were 31.21 mm caudally and 14.54 mm cranially; underscanning occurred in 4.36% (caudal) and 0.89% (cranial). The tool enables objective, large-scale monitoring of LDCT quality—reducing routine manual workload through exception-based human oversight, flagging protocol deviations, and supporting cross-center benchmarking—and may facilitate dose optimization by reducing systematic over- and underscanning. Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

► Show Figures

Figure 1

13 pages, 2459 KB

Open AccessEditor’s ChoiceArticle

Visual Large Language Models in Radiology: A Systematic Multimodel Evaluation of Diagnostic Accuracy and Hallucinations

by Marc Sebastian von der Stück, Roman Vuskov, Simon Westfechtel, Robert Siepmann, Christiane Kuhl, Daniel Truhn and Sven Nebelung

Life 2026, 16(1), 66; https://doi.org/10.3390/life16010066 - 1 Jan 2026

Viewed by 1296

Abstract

Visual large language models (VLLMs) are discussed as potential tools for assisting radiologists in image interpretation, yet their clinical value remains unclear. This study provides a systematic and comprehensive comparison of general-purpose and biomedical VLLMs in radiology. We evaluated 180 representative clinical images with validated reference diagnoses (radiography, CT, MRI; 60 each) using seven VLLMs (ChatGPT-4o, Gemini 2.0, Claude Sonnet 3.7, Perplexity AI, Google Vision AI, LLaVA-1.6, LLaVA-Med-v1.5). Each model interpreted the image without and with clinical context. Mixed-effects logistic regression models assessed the influence of model, modality, and context on diagnostic performance and hallucinations (fabricated findings or misidentifications). Diagnostic accuracy varied significantly across all dimensions (p ≤ 0.001), ranging from 8.1% to 29.2% across models, with Gemini 2.0 performing best and LLaVA performing weakest. CT achieved the best overall accuracy (20.7%), followed by radiography (17.3%) and MRI (13.9%). Clinical context improved accuracy from 10.6% to 24.0% (p < 0.001) but shifted the model to rely more on textual information. Hallucinations were frequent (74.4% overall) and model-dependent (51.7–82.8% across models; p ≤ 0.004). Current VLLMs remain diagnostically unreliable, heavily context-biased, and prone to generating false findings, which limits their clinical suitability. Domain-specific training and rigorous validation are required before clinical integration can be considered. Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

► Show Figures

Figure 1

15 pages, 867 KB

Open AccessArticle

Neonatal and Birth Risk Factors for Type 1 Diabetes Mellitus: Prediction Using an Artificial Neural Network

by Claudiu Cobuz, Mădălina Ungureanu-Iuga and Maricela Cobuz

Life 2025, 15(12), 1800; https://doi.org/10.3390/life15121800 - 24 Nov 2025

Viewed by 567

Abstract

Type 1 Diabetes Mellitus (T1D) can be related to various factors, including neonatal and perinatal conditions. This study investigated the impact of neonatal and perinatal factors—Apgar score, birth weight, feeding type, sex, and delivery type—on the risk of Type 1 Diabetes Mellitus and evaluated predictive models. A cohort of 327 patients was analyzed using correlations, General Linear Model, and artificial neural network. T1D patients showed higher birth weight, lower Apgar score, a predominance of formula feeding, and more cesarean deliveries. Diabetes risk showed a moderate positive correlation with birth weight class and nutrition type (p < 0.05) and a weak negative correlation with Apgar score (p < 0.05), while birth weight, birth weight class, and nutrition type were all weakly positively correlated with each other and with delivery type (p < 0.05). General linear model identified nutrition type, birth weight, and Apgar score as key predictors, with significant interactions among them (R² = 0.88). Artificial neural network achieved high accuracy (84.2%), AUC (0.95), sensitivity (89.1%), and specificity (76.7%). ANN successfully modeled the complex non-linear interactions among early-life factors, allowing it to discriminate between high-risk and low-risk cases, as evidenced by the low prediction errors (RMSE 0.31 and MAE 0.09) and strong agreement (Kappa 0.81). The models’ strong internal predictive performance points to potential applications in early T1D diagnosis and personalized management, although confirmation in larger and independent datasets is needed. Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

► Show Figures

Figure 1

28 pages, 24418 KB

Open AccessArticle

PICU Face and Thoracoabdominal Detection Using Self-Supervised Divided Space–Time Mamba

by Mohamed Khalil Ben Salah, Philippe Jouvet and Rita Noumeir

Life 2025, 15(11), 1706; https://doi.org/10.3390/life15111706 - 4 Nov 2025

Viewed by 1114

Abstract

Non-contact vital sign monitoring in Pediatric Intensive Care Units is challenged by frequent occlusions, data scarcity, and the need for temporally stable anatomical tracking to extract reliable physiological signals. Traditional detectors produce unstable tracking, while video transformers are too computationally intensive for deployment on resource-limited clinical hardware. We introduce Divided Space–Time Mamba, an architecture that decouples spatial and temporal feature learning using State Space Models to achieve linear-time complexity, over 92% lower than standard transformers. To handle data scarcity, we employ self-supervised pre-training with masked autoencoders on over 50 k domain-specific video clips and further enhance robustness with multimodal RGB-D input. Our model demonstrates superior performance, achieving 0.96 mAP@0.5, 0.62 mAP50-95, and 0.95 rotated IoU. Operating at 23 FPS (43 ms latency), our method is approximately 1.9× faster than VideoMAE and 5.7× faster than frame-wise YOLOv8, demonstrating its suitability for real-time clinical monitoring. Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

► Show Figures

Figure 1

14 pages, 3065 KB

Open AccessArticle

Artificial Intelligence (AI) for Programmed Death Ligand-1 (PD-L1) Immunohistochemical Assessment in Urothelial Carcinomas: “Teaching” Cell Differentiation to AI Systems

by Ioan Alin Nechifor-Boilă, Adela Nechifor-Boilă, Andrada Loghin, Carmen Mihaela Mihu, Carmen Stanca Melincovici, Mădălin Mihai Onofrei, Călin Bogdan Chibelean, Orsolya Martha and Angela Borda

Life 2025, 15(6), 839; https://doi.org/10.3390/life15060839 - 22 May 2025

Cited by 1 | Viewed by 1585 | Correction

Abstract

Assessment of Programmed Death-Ligand 1 (PD-L1) immunohistochemical (IHC) expression on tumor cells (TCs) and immune cells (ICs) in bladder cancer (BC) is challenging. Artificial Intelligence (AI) has potential for accurate PD-L1 IHC scoring, but its efficiency remains debatable. Our aim was to compare two AI protocols provided by the free QuPath software (v0.5.1) (Selected Area Interpretation (AI-SAI) and Whole Slide Imaging (AI-WSI)) with manual PD-L1 IHC scoring. A total of 43 BCs were included. PD-L1 IHC was performed using the SP263 clone. The IHC slides were digitized and further imported into QuPath. The PD-L1 positivity threshold was set at 25%. Statistically significant correlations were observed between AI-SAI and manual interpretation for both TCs (r = 0.85) and ICs (r = 0.57). AI-WSI yielded comparable results, with correlation coefficients of r = 0.82 for TCs and r = 0.56 for ICs. However, AI-SAI demonstrated stronger agreement with manual assessment (κ = 0.86) compared to AI-WSI (κ = 0.65). Receiver Operating Characteristic (ROC) analysis further supported the superiority of AI-SAI, with higher AUC values for both TCs (0.96 vs. 0.92) and ICs (0.92 vs. 0.90). Our findings indicate that AI-SAI is preferable to AI-WSI, particularly in BC cases with high PD-L1-positive TC content. Nevertheless, supervision by an experienced pathologist is mandatory. Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

► Show Figures

Graphical abstract

14 pages, 4139 KB

Open AccessArticle

Artificial Intelligence in the Surgery-First Approach: Harnessing Deep Learning for Enhanced Condylar Reshaping Analysis: A Retrospective Study

by Umberto Committeri, Gabriele Monarchi, Massimiliano Gilli, Angela Rosa Caso, Federica Sacchi, Vincenzo Abbate, Stefania Troise, Giuseppe Consorti, Francesco Giovacchini, Valeria Mitro, Paolo Balercia and Antonio Tullio

Life 2025, 15(2), 134; https://doi.org/10.3390/life15020134 - 21 Jan 2025

Cited by 7 | Viewed by 2724

Abstract

Background: The surgery-first approach (SFA) in orthognathic surgery eliminates the need for pre-surgical orthodontic treatment, significantly reducing overall treatment time. However, reliance on a compromised occlusion introduces risks of condylar displacement and remodeling. This study employs artificial intelligence (AI) and deep learning to analyze condylar behavior, comparing the outcomes of SFA to the traditional surgery-late approach (SLA). Methods: A retrospective analysis was conducted on 77 patients (18 SFA and 59 SLA) treated at Perugia Hospital between 2016 and 2022. Preoperative (T0) and 12-month postoperative (T1) cone-beam computed tomography (CBCT) scans were analyzed using the 3D Slicer software and its Dental Segmentator extension, powered by a convolutional neural network (CNN). This automated approach reduced segmentation time from 7 h to 5 min. Pre- and postoperative 3D models were compared to assess linear and rotational deviations in condylar morphology, stratified via dentoskeletal classification and surgical techniques. Results: Both the SFA and SLA achieved high surgical accuracy (<2 mm linear deviation and <2° rotational deviation). The SFA and SLA exhibited similar rates of condylar surface remodeling, with minor differences in resorption and formation across dentoskeletal classifications. Mean surface changes were 0.41 mm (SFA) and 0.36 mm (SLA, p < 0.05). Conclusions: Deep learning enables rapid, precise CBCT analysis and shows promise for the early detection of condylar changes. The SFA does not increase adverse effects on condylar morphology compared to SLA, supporting its safety and efficacy when integrated with AI technologies. Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

► Show Figures

Figure 1

Review

Jump to: Research, Other

26 pages, 1131 KB

Open AccessReview

Artificial Intelligence-Powered Quality Assurance: Transforming Diagnostics, Surgery, and Patient Care—Innovations, Limitations, and Future Directions

by Yoojin Shin, Mingyu Lee, Yoonji Lee, Kyuri Kim and Taejung Kim

Life 2025, 15(4), 654; https://doi.org/10.3390/life15040654 - 16 Apr 2025

Cited by 14 | Viewed by 6697

Abstract

Artificial intelligence is rapidly transforming quality assurance in healthcare, driving advancements in diagnostics, surgery, and patient care. This review presents a comprehensive analysis of artificial intelligence integration—particularly convolutional and recurrent neural networks—across key clinical domains, significantly enhancing diagnostic accuracy, surgical performance, and pathology evaluation. Artificial intelligence-based approaches have demonstrated clear superiority over conventional methods: convolutional neural networks achieved 91.56% accuracy in scanner fault detection, surpassing manual inspections; endoscopic lesion detection sensitivity rose from 2.3% to 6.1% with artificial intelligence assistance; and gastric cancer invasion depth classification reached 89.16% accuracy, outperforming human endoscopists by 17.25%. In pathology, artificial intelligence achieved 93.2% accuracy in identifying out-of-focus regions and an F1 score of 0.94 in lymphocyte quantification, promoting faster and more reliable diagnostics. Similarly, artificial intelligence improved surgical workflow recognition with over 81% accuracy and exceeded 95% accuracy in skill assessment classification. Beyond traditional diagnostics and surgical support, AI-powered wearable sensors, drug delivery systems, and biointegrated devices are advancing personalized treatment by optimizing physiological monitoring, automating care protocols, and enhancing therapeutic precision. Despite these achievements, challenges remain in areas such as data standardization, ethical governance, and model generalizability. Overall, the findings underscore artificial intelligence’s potential to outperform traditional techniques across multiple parameters, emphasizing the need for continued development, rigorous clinical validation, and interdisciplinary collaboration to fully realize its role in precision medicine and patient safety. Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

► Show Figures

Figure 1

Other

Jump to: Research, Review

12 pages, 3027 KB

Open AccessCase Report

New Insights into Molecular Mechanisms and Radiomics in Non-Contrast CT for Aortic Dissection: A Case Report and Literature Review

by Jian-Cheng Tian, Jia-Hao Zhou, Jui-Yuan Chung, Po-Chen Lin, Giou-Teng Yiang, Ya-Chih Yang and Meng-Yu Wu

Life 2026, 16(1), 14; https://doi.org/10.3390/life16010014 - 22 Dec 2025

Viewed by 899

Abstract

Background: Computed tomography (CT) angiography is widely regarded as the gold standard for diagnosing acute aortic dissection. However, in patients with contraindications to iodinated contrast media, such as those with renal insufficiency or hemodynamic instability, non-contrast CT may offer a viable alternative for initial evaluation. Understanding the molecular mechanisms underlying aortic dissection, including extracellular matrix degradation, smooth muscle cell apoptosis, and inflammatory pathways, is crucial for developing novel diagnostic and therapeutic approaches. This report describes a single case of acute Stanford type A aortic dissection initially detected on non-contrast CT. Case Presentation: We describe a 74-year-old man who presented to the emergency department with fever and suspected infection, but without chest pain. An incidental finding on non-contrast CT revealed ascending aortic dilatation, pericardial effusion, and a suspected intimal flap. Subsequent CT angiography confirmed a Stanford type A aortic dissection. Conclusions: This case highlights the potential value of non-contrast CT in the early detection of aortic dissection, particularly when CT angiography cannot be performed. Recent advances in artificial intelligence (AI) and radiomic analysis have shown promise in augmenting the diagnostic capabilities of non-contrast CT by identifying subtle imaging features that may correlate with underlying molecular pathology and elude human observers. Emerging evidence suggests that radiomic features may reflect molecular alterations in the aortic wall, including metalloproteinase activity, collagen degradation, and inflammatory cell infiltration. Incorporating AI-assisted interpretation alongside insights into molecular mechanisms could facilitate earlier diagnosis, improve risk stratification, and guide personalized treatment strategies in critically ill patients. Although non-contrast CT has limited sensitivity for aortic dissection, it may still reveal crucial findings in selected cases and should be considered when contrast-enhanced imaging is not feasible. Ongoing progress in AI, radiomics, and molecular biomarker research may further expand the clinical applications of non-contrast CT in emergency cardiovascular care and bridge the gap between imaging phenotypes and molecular endotypes. These findings are hypothesis-generating and require validation in larger cohorts before clinical generalization. Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

► Show Figures

Figure 1

1 pages, 158 KB

Open AccessCorrection

Correction: Nechifor-Boilă et al. Artificial Intelligence (AI) for Programmed Death Ligand-1 (PD-L1) Immunohistochemical Assessment in Urothelial Carcinomas: “Teaching” Cell Differentiation to AI Systems. Life 2025, 15, 839

Life 2025, 15(8), 1297; https://doi.org/10.3390/life15081297 - 15 Aug 2025

Viewed by 605

Abstract

In the published publication [...] Full article

(This article belongs to the Special Issue Current and Future Perspectives of Artificial Intelligence in Medicine)

Show export options Show export options

Select all

Export citation of selected articles as:

Displaying articles 1-9

Journal Menu

Journal Browser

Current and Future Perspectives of Artificial Intelligence in Medicine

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (9 papers)

Research

Review

Other

Further Information

Guidelines

MDPI Initiatives

Follow MDPI