Next Article in Journal
Treatment of Molar-Incisor Hypomineralization and Its Impact on Child and Adolescent Oral Health: A Comprehensive Bibliometric Analysis
Previous Article in Journal
A Deep-Learning-Based Dynamic Multidimensional Memory-Augmented Personalized Recommendation Research
Previous Article in Special Issue
Stability of Feature Selection in Multi-Omics Data Analysis
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Editorial

Recent Progress and Challenges of Artificial Intelligence in Bioinformatics and New Medicine

1
School of Computer Science, Northwestern Polytechnical University, 1 Dongxiang Rd., Xi’an 710072, China
2
Key Laboratory of Big Data Storage and Management, Ministry of Industry and Information Technology, Northwestern Polytechnical University, 1 Dongxiang Rd., Xi’an 710072, China
*
Authors to whom correspondence should be addressed.
Appl. Sci. 2025, 15(17), 9598; https://doi.org/10.3390/app15179598 (registering DOI)
Submission received: 20 August 2025 / Accepted: 28 August 2025 / Published: 31 August 2025

1. Introduction

The exponential growth of big data in biology, medical science, and public health is fundamentally transforming the landscape of biomedical research and therapeutic development [1]. This data-driven revolution, accelerated by disruptive technologies in life and health sciences, has positioned artificial intelligence (AI) as a strategic cornerstone for advancing novel medicine and understanding complex disease mechanisms [2,3]. Cross-disciplinary convergence—spanning life sciences, clinical medicine, and computational informatics—is catalyzing unprecedented innovations in bioinformatics and translational research [4,5,6,7].
Significant progress has been made in leveraging AI for critical biomedical challenges. Deep learning architectures now enable precise analysis of multi-modal biological data [8], while network-based approaches decode intricate gene interactions underlying pathologies like cancer [9,10,11]. Recent advances include in silico modeling of cellular processes [12], AI-enhanced histopathological quantification [13], and personalized health monitoring through wearable biosignal interpretation [14]. Concurrently, large-scale omics integration—particularly single-cell sequencing—has unveiled unprecedented resolution of cellular heterogeneity and microenvironment dynamics [15,16,17].
However, this rapid progression faces persistent challenges. Key among these are the complexities of multi-omics data integration, the curse of dimensionality in high-feature/small-sample scenarios, and the demand for clinically interpretable AI solutions [18,19,20]. Data modality heterogeneity (e.g., genomic, imaging, wearable biosensors) further complicates model generalizability [21]. In histopathology, subjective manual analysis remains burdensome despite computer vision advancements [22], while in genomics, feature selection instability across omics layers undermines biomarker reliability [23]. Translational hurdles also persist, particularly in scaling advanced therapies like cell/gene treatments through constrained clinical infrastructures [24].
Addressing these frontiers necessitates the development of advanced computational methodologies that integrate innovation with tangible biomedical impact. Recent advances demonstrate how tailored AI frameworks—including ensemble deep learning architectures, self-supervised personalization techniques, and novel network algorithms—are overcoming domain-specific barriers by enabling transformative capabilities across multiple dimensions. These approaches facilitate the discovery of molecular biomarkers through integrative multi-omics analysis, automate the transformation of complex histopathological patterns into quantitative diagnostic criteria, enhance interpretation of multimodal physiological signals for personalized health monitoring, and optimize generative models for scalable biological simulation. Collectively, these methodologies exemplify AI’s capacity to distill actionable insights from biomedical complexity, driving innovations across diagnostics, therapeutic development, and precision intervention strategies. This paradigm shift is fundamentally reshaping computational medicine workflows while establishing robust pipelines for clinical translation.

2. An Overview of Published Articles

Recent research highlights significant advancements in artificial intelligence across diverse biomedical domains, enhancing precision, efficiency, and inclusivity.
In the critical field of cancer prognostics, Xu et al. introduced a significant advance with EP-WGCNA [25]. This weighted gene co-expression network employs a novel Euclidean-Pearson approach to identify key ferroptosis-related genes (FRGs) associated with gastric cancer. Their resulting prognostic model demonstrated substantially improved survival prediction capabilities compared to conventional methods. The clinical relevance of these identified biomarkers was further supported by experimental validation and pathway enrichment analysis.
When investigating challenges in biomarker stability, Łukaszuk et al. analyzed feature selection consistency in multi-omics data for TP53 mutation prediction—mutations associated with poor clinical outcomes in cancer [26]. Using three classifiers with L1 regularization and five-fold cross-validation across 15 TCGA datasets, they measured stability via the Nogueira metric. Their findings revealed that higher regularization strength consistently produced optimal stability across all classifiers and omics layers. Significantly, miRNA data layers exhibited the highest overall stability. In contrast, mutation and RNA expression layers showed substantially lower stability, particularly when regularization was weak. This research underscores the critical impact of data type, regularization strength, and careful analytical validation on the reliability of biomarker discovery in high-dimensional multi-omics data.
Advancing reproductive medicine, Raudonis et al. [27] developed an ensemble AI framework (UNet++, UNet, ResNet34-UNet) to determine the receptivity stage of the endometrium in HE-stained histology images. Their technique integrated image segmentation with statistical analysis of stromal nuclear feature evolution. Demonstrating superior segmentation performance (Dice: 0.95, IoU: 0.90), the approach revealed statistically significant differences in nuclear count and stromal tissue size across pre-receptive, receptive, and post-receptive stages. This demonstrates promising potential to enhance fertility treatments through objective endometrial tissue analysis.
Advancing personalized health monitoring, Islam et al. [28] leveraged wearable technology for stress prediction. Their approach employs self-supervised pre-training on 1D CNNs to develop personalized models using electrodermal activity data. Remarkably, on the WESAD dataset, their method achieves baseline performance with less than 30% labeled data, enabling real-time precision health interventions despite the inherent challenges of subjective biosignals.
In the field of computer vision, Wang et al. introduced SSSGAN, a generative approach that significantly enhances computational efficiency in semantic image composition by utilizing a single-generator design [29]. This method overcomes the scalability limitation of traditional multi-generator approaches, reducing the number of required generators to a constant while maintaining competitive performance. Complementing advancements in medical image analysis, Gonciar et al. developed a semi-automatic method for quantifying myocardial fibrosis in cardiac histopathology using objective digital color analysis (CIELAB-based methods) [30]. This approach minimizes subjectivity inherent in visual assessment and demonstrates comparable performance to established analytical techniques like stereology and ImageJ (version 1.54).
Separately, Kim et al. applied an ensemble of VGGish and YAMNet to Speech Emotion Recognition (SER) [31]. Using STFT preprocessing, speech was converted to spectrograms, segmented, and represented as Gaussian distributions, with data quality assessed by correlation coefficients. This reduced data scale and minimized uncertainty. Explanatory techniques (Grad CAM, LIME, and occlusion sensitivity) provided interpretable insights. The ensemble model achieved 87% accuracy across three datasets, outperforming individual models and showing adaptability to diverse environments. Key emotional areas identified by Grad CAM were converted to audio for analysis, enhancing reliability and explainability.
Similarly employing a comprehensive approach to complex system optimization, Aguilar-Gallardo et al. examined hospital-integrated GMP facilities for advanced therapies (ATMPs) [32]. While presenting promising treatment options, this integration demands significant adaptation within constrained healthcare infrastructures. Their analysis outlines critical implementation strategies, emphasizing enabling technologies like PAT, continuous manufacturing, and AI-driven process control. Success demands tailored quality systems prioritizing product quality over compliance alone; substantial investment; and sustained multidisciplinary collaboration bridging research and clinical impact.
Collectively, these diverse studies powerfully illustrate AI’s transformative role in driving precision, efficiency, and inclusivity across the biomedical landscape. Navigating the accompanying ethical and translational hurdles will require sustained collaborative validation efforts to maximize the clinical impact of these innovations and ultimately improve patient outcomes.

3. Conclusions

These studies highlight significant advancements in biomedical artificial intelligence, particularly in precision diagnostics, personalized health monitoring, and therapeutic innovation. Breakthroughs in computational methodologies—such as gene co-expression networks, self-supervised learning for biosignals, GAN-based architectures, and ensemble models for histopathology—demonstrate potential to enhance precision, scalability, and efficiency in tackling complex biomedical challenges. These findings underscore the necessity of interdisciplinary convergence to navigate data heterogeneity, ensure biomarker stability, and achieve robust clinical translation.
Future research should prioritize ethical frameworks, validation across diverse populations, and infrastructure investments to address challenges in data integration, model interpretability, and clinical implementation.

Author Contributions

T.W. and X.Z.: writing—original draft preparation; T.W., Y.W. and J.P.: writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

This work has been supported by the National Natural Science Foundation of China (grant number: 62102319).

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Ashley, E.A. Towards precision medicine. Nat. Rev. Genet. 2016, 17, 507–522. [Google Scholar] [CrossRef]
  2. He, J.; Baxter, S.L.; Xu, J.; Xu, J.; Zhou, X.; Zhang, K. The practical implementation of artificial intelligence technologies in medicine. Nat. Med. 2019, 25, 30–36. [Google Scholar] [CrossRef]
  3. Zhou, H.Y.; Li, Y.; Li, J.; Meng, J.; Wu, A. Unleashing the potential of artificial intelligence in infectious diseases. Natl. Sci. Rev. 2025, 12, nwaf004. [Google Scholar] [CrossRef]
  4. Patel, A.U.; Gu, Q.; Esper, R.; Maeser, D.; Maeser, N. The crucial role of interdisciplinary conferences in advancing explainable AI in healthcare. BioMedInformatics 2024, 4, 1363–1383. [Google Scholar] [CrossRef]
  5. Wang, T.; Luo, Z. Large language models transform biological research: From architecture to utilization. Sci. China Inf. Sci. 2025, 68, 170101. [Google Scholar] [CrossRef]
  6. Jiang, T.; Cao, S.; Liu, Y.; Zhang, Z.; Liu, B.; Luo, R.; Wang, G.; Wang, Y. cuteFC: Regenotyping structural variants through an accurate and efficient force-calling method. Genome Biol. 2025, 26, 166. [Google Scholar] [CrossRef] [PubMed]
  7. Wang, H.; Yang, Y.; Huang, Y.; Zang, T.; Liu, Y. TDLM: A Diffusion Language Model for TCR Sequence Exploration and Generation. In Proceedings of the 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Lisboa, Portugal, 3–6 December 2024; pp. 113–120. [Google Scholar]
  8. Acosta, J.N.; Falcone, G.J.; Rajpurkar, P.; Topol, E.J. Multimodal biomedical AI. Nat. Med. 2022, 28, 1773–1784. [Google Scholar] [CrossRef]
  9. Ranea, J.A.G.; Perkins, J.; Chagoyen, M.; Díaz-Santiago, E.; Pazos, F. Network-based methods for approaching human pathologies from a phenotypic point of view. Genes 2022, 13, 1081. [Google Scholar] [CrossRef] [PubMed]
  10. Zhang, L.; Wang, S.; Wang, Y.; Zhao, T.; Wren, J. HBFormer: A single-stream framework based on hybrid attention mechanism for identification of human-virus protein–protein interactions. Bioinformatics 2024, 40, btae724. [Google Scholar] [CrossRef] [PubMed]
  11. Song, W.; Xu, L.; Han, C.; Tian, Z.; Zou, Q.; Gao, X. Drug–target interaction predictions with multi-view similarity network fusion strategy and deep interactive attention mechanism. Bioinformatics 2024, 40, btae346. [Google Scholar] [CrossRef]
  12. Barh, D.; Yiannakopoulou, E.C.; Salawu, E.O.; Bhattacharjee, A.; Chowbina, S.; Nalluri, J.J.; Ghosh, P.; Azevedo, V. In silico disease model: From simple networks to complex diseases. In Animal Biotechnology; Academic Press: Cambridge, MA, USA, 2020; pp. 441–460. [Google Scholar]
  13. Ma, Y.; Jamdade, S.; Konduri, L.; Sailem, H. AI in Histopathology Explorer for comprehensive analysis of the evolving AI landscape in histopathology. npj Digit. Med. 2025, 8, 156. [Google Scholar] [CrossRef]
  14. Vaghasiya, J.V.; Mayorga-Martinez, C.C.; Pumera, M. Wearable sensors for telehealth based on emerging materials and nanoarchitectonics. npj Flex. Electron. 2023, 7, 26. [Google Scholar] [CrossRef]
  15. Wu, X.; Yang, X.; Dai, Y.; Zhao, Z.; Zhu, J.; Guo, H.; Yang, R. Single-cell sequencing to multi-omics: Technologies and applications. Biomark. Res. 2024, 12, 110. [Google Scholar] [CrossRef]
  16. Chen, C.; Wang, J.; Pan, D.; Wang, X.; Xu, Y.; Yan, J.; Wang, L.; Yang, X.; Yang, M.; Liu, G. Applications of multi-omics analysis in human diseases. MedComm 2023, 4, e315. [Google Scholar] [CrossRef]
  17. Zhu, P.; Shu, H.; Wang, Y.; Wang, X.; Zhao, Y.; Hu, J.; Peng, J.; Shang, X.; Tian, Z.; Chen, J.; et al. MAEST: Accurately spatial domain detection in spatial transcriptomics with graph masked autoencoder. Brief. Bioinform. 2025, 26, bbaf086. [Google Scholar] [CrossRef]
  18. Ballard, J.L.; Wang, Z.; Li, W.; Shen, L.; Long, Q. Deep learning-based approaches for multi-omics data integration and analysis. BioData Min. 2024, 17, 38. [Google Scholar] [CrossRef] [PubMed]
  19. Hagos, D.H.; Aryal, S.K.; Ymele-Leki, P.; Burge, L.L. AI-driven multimodal colorimetric analytics for biomedical and behavioral health diagnostics. Comput. Struct. Biotechnol. J. 2025, 27, 2219–2232. [Google Scholar] [CrossRef] [PubMed]
  20. Zhang, X.; Chen, J.; Wang, Y.; Wang, X.; Hu, J.; Peng, J.; Shang, X.; Wang, Y.; Wang, T. cfMethylPre: Deep transfer learning enhances cancer detection based on circulating cell-free DNA methylation profiling. Brief. Bioinform. 2025, 26, bbaf303. [Google Scholar] [CrossRef] [PubMed]
  21. Mezei, T.; Kolcsár, M.; Joó, A.; Gurzu, S. Image analysis in histopathology and cytopathology: From early days to current perspectives. J. Imaging 2024, 10, 252. [Google Scholar] [CrossRef]
  22. Ma, D.; Fan, C.; Sano, T.; Kawabata, K.; Nishikubo, H.; Imanishi, D.; Sakuma, T.; Maruo, K.; Yamamoto, Y.; Matsuoka, T.; et al. Beyond Biomarkers: Machine Learning-Driven Multiomics for Personalized Medicine in Gastric Cancer. J. Pers. Med. 2025, 15, 166. [Google Scholar] [CrossRef]
  23. Austin, C.P. Opportunities and challenges in translational science. Clin. Transl. Sci. 2021, 14, 1629–1647. [Google Scholar] [CrossRef]
  24. Luo, X.; Wang, Y.; Zou, Q.; Xu, L.; Ioshikhes, I. Recall DNA methylation levels at low coverage sites using a CNN model in WGBS. PLoS Comput. Biol. 2023, 19, e1011205. [Google Scholar] [CrossRef]
  25. Xu, Y.; Zhang, H.; Cao, D.; Ning, Z.; Zhu, L.; Liu, X. A Novel Prognostic Model for Gastric Cancer with EP_Dis-Based Co-Expression Network Analysis. Appl. Sci. 2023, 13, 7108. [Google Scholar] [CrossRef]
  26. Łukaszuk, T.; Krawczuk, J.; Żyła, K.; Kęsik, J. Stability of Feature Selection in Multi-Omics Data Analysis. Appl. Sci. 2024, 14, 11103. [Google Scholar] [CrossRef]
  27. Raudonis, V.; Bartasiene, R.; Minajeva, A.; Saare, M.; Drejeriene, E.; Kozlovskaja-Gumbriene, A.; Salumets, A. Towards metric-driven difference detection between receptive and nonreceptive endometrial samples using automatic histology image analysis. Appl. Sci. 2024, 14, 5715. [Google Scholar] [CrossRef]
  28. Islam, T.; Washington, P. Individualized stress mobile sensing using self-supervised pre-training. Appl. Sci. 2023, 13, 12035. [Google Scholar] [CrossRef]
  29. Wang, Z.; Liu, Z. Unlocking Efficiency in Fine-Grained Compositional Image Synthesis: A Single-Generator Approach. Appl. Sci. 2023, 13, 7587. [Google Scholar] [CrossRef]
  30. Gonciar, D.; Berciu, A.G.; Danku, A.E.; Lorenzovici, N.; Dulf, E.-H.; Mocan, T.; Nicula, S.-M.; Agoston-Coldea, L. The Quantification of Myocardial Fibrosis on Human Histopathology Images by a Semi-Automatic Algorithm. Appl. Sci. 2024, 14, 7696. [Google Scholar] [CrossRef]
  31. Kim, T.W.; Kwak, K.C. Speech emotion recognition using deep learning transfer models and explainable techniques. Appl. Sci. 2024, 14, 1553. [Google Scholar] [CrossRef]
  32. Aguilar-Gallardo, C.; Bonora-Centelles, A. Integrating artificial intelligence for academic advanced therapy medicinal products: Challenges and opportunities. Appl. Sci. 2024, 14, 1303. [Google Scholar] [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Wang, T.; Zhang, X.; Wang, Y.; Peng, J. Recent Progress and Challenges of Artificial Intelligence in Bioinformatics and New Medicine. Appl. Sci. 2025, 15, 9598. https://doi.org/10.3390/app15179598

AMA Style

Wang T, Zhang X, Wang Y, Peng J. Recent Progress and Challenges of Artificial Intelligence in Bioinformatics and New Medicine. Applied Sciences. 2025; 15(17):9598. https://doi.org/10.3390/app15179598

Chicago/Turabian Style

Wang, Tao, Xuchao Zhang, Yongtian Wang, and Jiajie Peng. 2025. "Recent Progress and Challenges of Artificial Intelligence in Bioinformatics and New Medicine" Applied Sciences 15, no. 17: 9598. https://doi.org/10.3390/app15179598

APA Style

Wang, T., Zhang, X., Wang, Y., & Peng, J. (2025). Recent Progress and Challenges of Artificial Intelligence in Bioinformatics and New Medicine. Applied Sciences, 15(17), 9598. https://doi.org/10.3390/app15179598

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop