Pediatric radiology is an inherently technology-dependent medical subspecialty [1]. In the past decades, medical imaging procedures in children increased substantially with the exception of a stable or decreasing number of computed tomography (CT) examinations, due to radiation protection efforts [2,3]. However, there is a significant lack of specialized, board-certified pediatric radiologists to cover the unique imaging requirements in children. The wide patient stature range from premature infants to potentially obese adolescents underlines the need for specific imaging expertise. For example, higher radiation doses have been reported for CT examinations of almost all body regions if children were examined in non-pediatric facilities [4]. In addition, reporting of pediatric imaging procedure results by adult/general radiologists or even non-radiology specialists (in particular during off-duty hours) might lead to wrong or delayed diagnoses [5,6,7]. Furthermore, novel techniques are developed for (and validated in) adult patient populations in the majority of cases. This means that they have to be adapted prior to their use in children and adolescents [8]. These facts imply the need for advancements in pediatric radiology to ensure optimal care for the smallest of our patients. If new scientific inputs emerge into clinical practice, we might be able to compensate at least partially for the lack of specialists.
A few years ago, we summarized modern developments in pediatric radiology, concentrating on CT dose reduction, advances in magnetic resonance imaging (MRI) techniques, and hybrid imaging such as positron emission tomography (PET)/MRI [9]. Nowadays, the clinical introduction of the photon-counting CT technology has emerged as a major step towards image quality optimization, for example, by applying multi-energy and fast-pitch scanning modes, virtual mono-energetic image reconstructions, or ultra-high resolution scans [10]. In addition, it offers a large potential for dose reduction, especially in pediatric patients [11]. It has been applied to different disorders, including, but not limited to, cardiovascular disease [12,13], lung pathologies [14], neurological disorders [15], and abdominopelvic pathologies [16]. Thus, photon-counting CT marks a major technological advancement in CT imaging for pediatric patients, as it can improve image quality while reducing the radiation dose and/or amount of necessary contrast media [10].
A major contemporary challenge in almost all medical subspecialties, including pediatric radiology, is the evidence-based implementation of techniques using artificial intelligence (AI) [17]. As mentioned above, adult-trained deep learning models do not necessarily transfer to pediatric patient collectives without adaptations [18]. Still, these techniques are already used for different applications, which can be assorted to different groups. First, AI tools can assist the radiologist in making a diagnosis. Examples include bone age estimation [19] or fracture assessment [20]. Furthermore, AI can be used to optimize workflow, for example, by performing automated tumor segmentations [21], which would be a time-consuming task if performed manually by the radiologist. Finally, AI is already implemented into clinical routines as a tool to enhance image quality by reducing artifacts due to an acceleration of MRI pulse sequences [22]. Furthermore, AI might allow for a reduction in the amount of necessary contrast media [23] or PET tracers [24], which consequently lowers potential side effects or ionizing radiation.
While most medical deep learning models are specifically developed to aid in diagnostics of a disorder or for automated image analysis, other approaches aim to use publicly available AI tools such as ChatGPT, e.g., for pediatric fracture evaluation [25] or bone age estimation [26]. However, these approaches have to be followed with care, as recently demonstrated in a study showing a low sensitivity when using ChatGPT for the detection of subtle pneumothoraces in pediatric patients [27]. This implies that a cautious evaluation of new AI methods is necessary prior to their implementation into daily clinical routines.
Recent advances in the field of ultrasound as a point-of-care imaging modality head for improvements in diagnostics by adding quantitative information. For example, radiomics-based analyses of kidney sonograms might differentiate between acute and chronic renal disorders [28]. Another example of a quantitative parameter is organ stiffness measured by ultrasound elastography. This parameter might be exemplarily used to differentiate between different types of spleen pathology in children [29]. Both of the aforementioned examples underline the strive for the acquisition of parameters that exceed pure visual imaging analysis in order to enhance diagnostic possibilities.
While new techniques, as described above, have to be developed and implemented continuously, the standardization of established methods also remains an ongoing challenge in pediatric radiology. First, this applies to the methodological aspect of how image acquisition is practically performed and by whom. Exemplarily, for cross-sectional imaging of neonates and infants, potentially requiring anesthesia, alternative methods such as “feed-and-wrap” approaches [30] are still inconsistently applied [31]. Therefore, retrospective data analyses concerning the use of clinically well-established methods might generate scientific evidence for their utilization. Second, harmonization of imaging protocols, for example, MRI sequences and sequence parameters, among institutions worldwide should be a topic to advance on [32]. If imaging is harmonized across specialized pediatric radiology departments, it may be more readily transferable to hospitals or regions where dedicated pediatric radiologists are unavailable.
In summary, pediatric radiologists will have to stay at the front of new technological developments to be able to provide high-quality care for children and adolescents. Collecting new evidence concerning diagnostics of pediatric diseases by imaging will help to transfer scientific research into clinical practice.
Conflicts of Interest
The author declares no conflicts of interest.
Abbreviations
The following abbreviations are used in this manuscript:
| AI | Artificial Intelligence |
| CT | Computed Tomography |
| MRI | Magnetic Resonance Imaging |
| PET | Positron Emission Tomography |
References
- Kantarcı, M.; Aydın, S.; Oğul, H.; Kızılgöz, V. New imaging techniques and trends in radiology. Diagn. Interv. Radiol. 2025, 31, 505–517. [Google Scholar] [CrossRef] [Scilit]
- Smith-Bindman, R.; Kwan, M.L.; Marlow, E.C.; Theis, M.K.; Bolch, W.; Cheng, S.Y.; Bowles, E.J.A.; Duncan, J.R.; Greenlee, R.T.; Kushi, L.H.; et al. Trends in Use of Medical Imaging in US Health Care Systems and in Ontario, Canada, 2000–2016. JAMA 2019, 322, 843–856. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Marin, J.R.; Rodean, J.; Hall, M.; Alpern, E.R.; Aronson, P.L.; Chaudhari, P.P.; Cohen, E.; Freedman, S.B.; Morse, R.B.; Peltz, A.; et al. Trends in Use of Advanced Imaging in Pediatric Emergency Departments, 2009–2018. JAMA Pediatr. 2020, 174, e202209. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Strauss, K.J.; Somasundaram, E.; Sengupta, D.; Marin, J.R.; Brady, S.L. Radiation Dose for Pediatric CT: Comparison of Pediatric versus Adult Imaging Facilities. Radiology 2019, 291, 158–167. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Scaife, J.H.; Martinez, D.J.; Clinker, C.E.; Glasgow, S.L.; Kahan, A.M.; Roach, C.M.; Russell, K.W.; Swendiman, R.A. Discordance in CT interpretations between pediatric radiology overreads and community and adult radiologists reads following traumatic injury. J. Pediatr. Surg. 2026, 61, 162765. [Google Scholar] [CrossRef] [Scilit]
- Rotem-Grunbaum, B.; Scheuerman, O.; Tamary, O.; Lakovsky, Y.; Shkalim Zemer, V.; Goldberg, L.; Soffair, N.; Bulkowstein, Y.; Hendelsman, S.; Amarilyo, G.; et al. Pediatric chest radiograph interpretation in a real-life setting. Eur. J. Pediatr. 2024, 183, 4435–4444. [Google Scholar] [CrossRef] [Scilit]
- Eakins, C.; Ellis, W.D.; Pruthi, S.; Johnson, D.P.; Hernanz-Schulman, M.; Yu, C.; Kan, J.H. Second opinion interpretations by specialty radiologists at a pediatric hospital: Rate of disagreement and clinical implications. AJR Am. J. Roentgenol. 2012, 199, 916–920. [Google Scholar] [CrossRef] [Scilit]
- Shin, H.J.; Son, N.H.; Kim, M.J.; Kim, E.K. Diagnostic performance of artificial intelligence approved for adults for the interpretation of pediatric chest radiographs. Sci. Rep. 2022, 12, 10215. [Google Scholar] [CrossRef] [Scilit]
- Staatz, G.; Daldrup-Link, H.E.; Herrmann, J.; Hirsch, F.W.; Schäfer, J.F.; Seehofnerova, A.; Sorantin, E.; Theruvath, A.J.; Lollert, A. From Xrays to PET/MR, and then?—Future imaging in pediatric radiology. Rofo 2019, 191, 357–366. [Google Scholar] [CrossRef] [Scilit]
- Horst, K.K.; Cao, J.Y.; McCollough, C.H.; El-Ali, A.; Frush, D.P.; Siegel, M.J.; Ramirez-Giraldo, J.C.; O’Donnell, T.; Bache, S.; Yu, L. Multi-institutional Protocol Guidance for Pediatric Photon-counting CT. Radiology 2024, 311, e231741. [Google Scholar] [CrossRef] [Scilit]
- Habib Geryes, B.; Moalla, S.; De Saint Etienne, E.; Moussier, A.; Pontana, F.; Boddaert, N.; Pontvianne, V.; Dangouloff-Ros, V. Photon-counting CT in pediatric patients: A multi-institutional evaluation of protocols and radiation exposure in 4,772 patients. Diagn. Interv. Imaging 2025, in press. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Hellms, S.; Werncke, T.; Böttcher, J.; Happel, C.M.; Eckstein, J.; Krueger, M.B.; Panknin, C.; Pfeil, A.; Kaireit, T.F.; Beerbaum, P.; et al. Reduction of radiation exposure and preserved image quality using photon-counting detector cardiac computed tomography without electrocardiographic gating in children with congenital heart disease. Eur. Radiol. 2026, 36, 324–333. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Lollert, A.; Souschek, F.; Abu-Tair, T.; Dette, F.; Duerr, G.D.; Kampmann, C.; Bäuerle, T.; Staatz, G. Photon-Counting CT Enables Higher Image Quality at a Lower Radiation Dose Compared with a 256-Slice Energy-Integrating Detector CT in Pediatric Patients with Congenital Heart Disease. Diagnostics 2026, 16, 735. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Krueger, M.B.; Werncke, T.; Eicke, M.; Schwerk, N.; Eckstein, J.; Huisinga, C.; Panknin, C.; Shin, H.O.; Silchmüller, F.J.; Schultze-Florey, R.E.; et al. Photon-counting Detector CT Enables Pediatric Low-dose Chest Imaging with Further Reduction of Radiation Exposure. Investig. Radiol. 2025. [Google Scholar] [CrossRef] [Scilit]
- Naceanceno, K.; Thomas, A.; Zhang, Z.; Sanchez, A.; Lappe, C.; Skinner, N.; Abadia, A.; Landrum, K.; McKinstry, R. Quantifying performance enhancements in photon-counting pediatric head CT: A direct comparison with energy-integrating CT. Pediatr. Radiol. 2025, 55, 1690–1701. [Google Scholar] [CrossRef] [Scilit]
- Siegel, M.J.; Thomas, M.A.; Haq, A.; Seymore, N.; Sodhi, K.S.; Abadia, A. Comparison of Radiation Dose and Image Quality in Pediatric Abdominopelvic Photon-Counting Versus Energy-Integrating Detector CT. J. Comput. Assist. Tomogr. 2026, 50, 73–80. [Google Scholar] [CrossRef] [Scilit]
- Kapur, J.; Kelly, B.S.; Vega, R. A modern radiologist’s guide to artificial intelligence. Pediatr. Radiol. 2026, 1–13. [Google Scholar] [CrossRef] [Scilit]
- Laborie, L.B.; Lee, J.; Antram, E.; Lein, R.K.; Shelmerdine, S.C. Performance of adult-trained artificial intelligence models in paediatric imaging-a scoping review. Eur. Radiol. 2026, 1–24. [Google Scholar] [CrossRef] [Scilit]
- Yuan, W.; Fan, P.; Zhang, L.; Pan, W.; Zhang, L. Bone Age Assessment Using Various Medical Imaging Techniques Enhanced by Artificial Intelligence. Diagnostics 2025, 15, 257. [Google Scholar] [CrossRef] [Scilit]
- Nguyen, T.; Maarek, R.; Hermann, A.L.; Kammoun, A.; Marchi, A.; Khelifi-Touhami, M.R.; Collin, M.; Jaillard, A.; Kompel, A.J.; Hayashi, D.; et al. Assessment of an artificial intelligence aid for the detection of appendicular skeletal fractures in children and young adults by senior and junior radiologists. Pediatr. Radiol. 2022, 52, 2215–2226. [Google Scholar] [CrossRef] [Scilit]
- Modanwal, G.; Kumar, S.; Viswanathan, V.; Morin, C.E.; Rees, M.A.; Squires, J.H.; Tang, E.R.; Katzenstein, H.M.; Towbin, A.J.; Madabhushi, A.; et al. Pediatric Personalized Deep Learning Models for Segmentation of Hepatoblastoma at CT and MRI. Radiol. Imaging Cancer 2026, 8, e250041. [Google Scholar] [CrossRef] [Scilit]
- Yoo, H.; Moon, H.E.; Kim, S.; Kim, D.H.; Choi, Y.H.; Cheon, J.E.; Lee, J.S.; Lee, S. Evaluation of Image Quality and Scan Time Efficiency in Accelerated 3D T1-Weighted Pediatric Brain MRI Using Deep Learning-Based Reconstruction. Korean J. Radiol. 2025, 26, 180–192. [Google Scholar] [CrossRef] [Scilit]
- Zhou, X.; Li, J.; Ke, T.; Xiong, D.; Liu, F.; Tan, N.; Duan, X.; Du, X.; Zhou, F.; Shen, W.; et al. Impact of contrast enhancement boost and super-resolution deep learning reconstruction on pediatric congenital heart disease CTA scans: Ultra-low contrast dose. BMC Med. Imaging 2025, 25, 477. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Vaccaro, M.; Rosa, E.; Placidi, E.; Guarnera, A.; Secinaro, A.; Gandolfo, C.; Garganese, M.C.; Napolitano, A. Deep learning for synthetic PET imaging: A systematic mapping review of techniques, metrics, and clinical relevance. Eur. Radiol. Exp. 2026, 10, 12. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Phua, J.K.; Tan, T.S.E. Flexing ChatGPT-4o’s Diagnostic Muscle: Detection of Fractures in the Ossifying Pediatric Elbow on Radiographs. Diagnostics 2025, 15, 2882. [Google Scholar] [CrossRef] [Scilit] [PubMed]
- Ozkan, E.; Koyun, M. Atlas-Assisted Bone Age Estimation from Hand-Wrist Radiographs Using Multimodal Large Language Models: A Comparative Study. Diagnostics 2026, 16, 487. [Google Scholar] [CrossRef] [Scilit]
- Wang, C.H.; Lin, P.C.; Shih, S.L.; Tsai, P.S.; Huang, W.H. Diagnostic Performance of ChatGPT-5 for Detecting Pediatric Pneumothorax on Chest Radiographs: A Multi-Prompt Evaluation. Diagnostics 2026, 16, 232. [Google Scholar] [CrossRef] [Scilit]
- De Leon-Benedetti, L.; Sultan, L.R.; Otero, H.J.; Morales-Tisnés, T.; Sims, J.; Fitzpatrick, K.; Fitzgerald, J.C.; Furth, S.; Laskin, B.L.; Viteri, B. Quantifying “Medical Renal Disease”: A Pediatric Pilot Study Using Ultrasound Radiomics for Differentiating Acute Kidney Injury and Chronic Kidney Disease. Diagnostics 2025, 15, 2112. [Google Scholar] [CrossRef] [Scilit]
- Cetin, T.; Tokur, O.; Bozkurt, H.B.; Aydin, S.; Memis, K.B.; Kantarci, M. Shear Wave Ultrasonographic Elastography in Pediatric Spleens and Its Role in Differential Diagnosis. Diagnostics 2024, 14, 1142. [Google Scholar] [CrossRef] [Scilit]
- Lollert, A.; Frey, K.S.; Hoffmann, C.; Herbst, M.; König, J.; Schmeh, I.; Dette, F.; Staatz, G. “Feed-and-wrap” technique versus deep sedation for neonatal magnetic resonance imaging: A retrospective comparative study. Eur. Radiol. 2024, 34, 7104–7114. [Google Scholar] [CrossRef] [Scilit]
- Hwang, M.; Barton, K.; Kim, J.S.; Machnitz, A.J.; Holm, T.; Maianski, I.; Setty, B.N.; Birkemeier, K.L.; Radhakrishnan, R.; Sreedher, G.; et al. Utilization of neonatal sedation and anesthesia: An SPR survey. Pediatr. Radiol. 2022, 52, 2630–2635. [Google Scholar] [CrossRef] [Scilit]
- Ferraciolli, S.F.; Boechat, M.I.; Shu, Y.; Anu, M.; Harris, C.; Van Vorstenbosch-Lynn, E.; Kilborn, T.; Lam, W.; Ho, M.L.; Kasznia-Brown, J.; et al. International standardization of pediatric magnetic resonance imaging protocols: Creation of the World Federation of Pediatric Imaging MR Protocols Committee. Pediatr. Radiol. 2025, 55, 375–383. [Google Scholar] [CrossRef] [Scilit]
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. |
© 2026 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.