Deep learning (DL)-based advancements in prostate cancer imaging: Artificial intelligence (AI)-based segmentation of ⁶⁸Ga-PSMSA PET for tumor volume assessment

Abstract

Positron emission tomography (PET) with gallium-68 prostate-specific membrane antigen (⁶⁸Ga-PSMA) has emerged as a promising imaging modality for evaluating prostate cancer (PC). Quantification of tumor volume is crucial for staging, radiotherapy treatment planning, response assessment, and prognosis in PC patients. This review provides an overview of the current methods and challenges in the assessment of regional and total tumor volumes using ⁶⁸Ga-PSMA PET. Traditional manual segmentation methods are time-consuming processes that are further challenged by inter-observer variability. Artificial intelligence (AI)-based segmentation techniques offer a promising solution to these challenges. AI algorithms, such as deep learning-based models, have shown remarkable performance in automating tumor segmentation tasks with high accuracy and efficiency. This review discusses the principles underlying AI-based segmentation algorithms, including convolutional neural networks, and their applications in PC imaging. Furthermore, the advantages of AI-based segmentation are highlighted, such as improved reproducibility, faster processing times, and potential for personalized medicine. Despite these advancements, AI-based segmentation faces significant challenges, including the need for standardized protocols, extensive validation studies, and seamless integration into clinical workflows. Addressing these limitations is essential for the widespread adoption of AI-based segmentation in ⁶⁸Ga-PSMA PET for PC, ultimately advancing the field and improving patient care.

Keywords

⁶⁸Ga-PSMA / PET-CT / prostate cancer / artificial intelligence / deep learning / radiotherapy / radiation treatment

References

[1] Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021; 71(3): 209-249.

[2] Beckham TH, Yang TJ, Gomez D, Tsai CJ. Metastasis-directed therapy for oligometastasis and beyond. Br J Cancer. 2021; 124(1): 136-141.

[3] Bluemel C, Krebs M, Polat B, et al. 68Ga-PSMA-PET/CT in patients with biochemical prostate cancer recurrence and negative 18F-choline-PET/CT. Clin Nucl Med. 2016; 41: 515-521.

[4] Asokendaran M, Meyrick DP, Skelly LA, Lenzo NP, Henderson A. Gallium-68 prostate-specific membrane antigen positron emission tomography/computed tomography compared with diagnostic computed tomography in relapsed prostate cancer. World J Nucl Med. 2019; 18(3): 232-237.

[5] Rao A, Vapiwala N, Schaeffer EM, Ryan C. Oligometastatic prostate cancer: A shrinking subset or an opportunity for cure? Am Soc Clin Oncol Educ Book. 2019; 39: 309-320.

[6] Afshar-Oromieh A, Zechmann CM, Malcher A, et al. Comparison of PET imaging with a (68)Ga-labelled PSMA ligand and (18)F-choline-based PET/CT for the diagnosis of recurrent prostate cancer. Nucl Med Mol Imaging. 2014; 41(1): 11-20.

[7] Sterzing F, Kratochwil C, Fiedler H, et al. (68)Ga-PSMA-11 PET/CT: a new technique with high potential for the radiotherapeutic management of prostate cancer patients. Eur J Nucl Med Mol Imaging. 2016; 43(1): 34-41.

[8] Kranzbühler B, Müller J, Becker AS, et al. Detection Rate and Localization of Prostate Cancer Recurrence Using 68Ga-PSMA-11 PET/MRI in Patients with Low PSA Values ≤ 0.5 ng/mL. J Nucl Med. 2020; 61(2): 194-201.

[9] Lindgren Belal S, Frantz S, Minarik D, et al. Applications of Artificial Intelligence in PSMA PET/CT for Prostate Cancer Imaging. Semin Nucl Med. 2024; 54(1): 141-149.

[10] Giesel FL, Hadaschik B, Cardinale J, et al. F-18 labelled PSMA-1007: biodistribution, radiation dosimetry and histopathological validation of tumor lesions in prostate cancer patients. Eur J Nucl Med Mol Imag. 2017; 44(4): 678-688.

[11] Lisney AR, Leitsmann C, Strauß A, Meller B, Bucerius JA, Sahlmann CO. The role of PSMA PET/CT in the primary diagnosis and follow-up of prostate cancer—a practical clinical review. Cancers. 2022; 14(15): 3638.

[12] Kopka K, Benesova M, Barinka C, et al. Glu-Ureido–Based inhibitors of prostate-specific membrane antigen: lessons learned during the development of a novel class of low molecular weight. theranostic radiotracers. J Nucl Med. 2017; 58(Suppl 2): 17S-26S.

[13] Moghadam SZ, Askari E, Aryana K. ^99mTc- PSMA left behind: a call for collaboration. Nucl Med Mol Imaging. 2022; 56(4): 218-219.

[14] Jadvar H, Calais J, Fanti S, et al. Appropriate use criteria for prostate-specific membrane antigen PET imaging. J Nucl Med. 2022; 63(1): 59-68.

[15] Tayara OM, Pełka K, Kunikowska J, et al. Comparison of Multiparametric MRI, [68Ga]Ga-PSMA-11 PET-CT, and Clinical Nomograms for Primary T and N Staging of Intermediate-to-High-Risk Prostate Cancer. Cancers (Basel). 2023; 15(24): 5838.

[16] Sharma AP, Kumar R, Chauhan R, et al. Accuracy of combined multi-parametric MRI and PSMA PET-CT in diagnosing localized prostate cancer: newer horizons for a biopsy-free pathway. Eur J Hybrid Imaging. 2023; 7(1): 24.

[17] Giunta EF, Brighi N, Gurioli G, Matteucci F, Paganelli G, De Giorgi U. 177Lu-PSMA therapy in metastatic prostate cancer: An updated review of prognostic and predictive biomarkers. Cancer Treat Rev. 2024; 125: 102699.

[18] Parghane RV, Basu S. PSMA-targeted radioligand therapy in prostate cancer: current status and future prospects. Expert Rev Anticancer Ther. 2023; 23(9): 959-975.

[19] Al-Ibraheem A, Abuhijla F, Salah S, et al. The influence of 68Ga-prostate-specific membrane antigen PET/computed tomography on prostate cancer staging and planning of definitive radiation therapy. Nucl Med Commun. 2021; 42(7): 811-817.

[20] Calais J, Kishan AU, Cao M, et al. Potential Impact of 68Ga-PSMA-11 PET/CT on the Planning of Definitive Radiation Therapy for Prostate Cancer. J Nucl Med. 2018; 59(11): 1714-1721.

[21] Roach PJ, Francis R, Emmett L, et al. The Impact of (68)Ga-PSMA PET/CT on Management Intent in Prostate Cancer: Results of an Australian Prospective Multicenter Study. J Nucl Med. 2018; 59(1): 82-88.

[22] Erickson BJ, Korfiatis P, Akkus Z, Kline TL. Machine Learning for Medical Imaging. Radiographics. 2017; 37(2): 505-515.

[23] Bagci U, Udupa JK, Yao J, Mollura DJ. Co-segmentation of functional and anatomical images. Med Image ComputComput Assist Interv. 2012; 15(Pt 3): 459-467.

[24] Zang S, Jiang C, Zhang L, et al. Deep learning based on 68Ga-PSMA-11 PET/CT for predicting pathological upgrading in patients with prostate cancer. Front Oncol. 2024; 13: 1273414.

[25] Trägårdh E, Borrelli P, Kaboteh R, et al. RECOMIA-acloud-based platform for artificial intelligenceresearch in nuclear medicine and radiology. EJNMMI Phys. 2020; 7(1): 51.

[26] Xie X, Zhou P, Li H, Lin Z, Yan S. Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models. IEEE Trans Pattern Anal Mach Intell. 2024; 46(12): 9508-9520.

[27] Khan AH, Pervez N, Usmani MS, et al. Intensifying the Hormonal and Radiation Treatment in Prostate Cancer Presenting with Bulky Pelvic Lymph Nodes: An Opportunity for a New Paradigm. Cureus. 2024; 16(1): e51518.

[28] Nikitas J, Lam E, Booker KA, et al. Randomized Trial of Prostate-Specific Membrane Antigen PET/CT Before Definitive Radiotherapy for Unfavorable Intermediate- and High-Risk Prostate Cancer (PSMA-dRT Trial). J Nucl Med. 2024; 65(7): 1076-1079.

[29] Fendler WP, Calais J, Eiber M, et al. Assessment of 68Ga-PSMA-11 PET Accuracy in Localizing Recurrent Prostate Cancer: A Prospective Single-Arm Clinical Trial. JAMA Oncol. 2019; 5(6): 856-863.

[30] Kendrick J, Francis RJ, Hassan GM, et al. Prognostic utility of RECIP 1.0 with manual and AI-based segmentations in biochemically recurrent prostate cancer from [⁶⁸Ga]Ga-PSMA-11 PET images. Eur J Nucl Med Mol Imaging. 2023; 50(13): 4077-4086.

RIGHTS & PERMISSIONS

2025 The Author(s). Precision Radiation Oncology published by John Wiley & Sons Australia, Ltd on behalf of Shandong Cancer Hospital & Institute.