Cancer continues to be one of the leading causes of mortality worldwide, with an estimated 19.3 million new cases and nearly 10 million cancer-related deaths reported globally each year [1]. In India, the National Cancer Registry Programme has projected a steady rise in cancer incidence, with an estimated burden exceeding 1.4 million new cases annually, compounded by a high proportion of patients presenting with locally advanced or metastatic disease [4]. A nationally representative mortality survey conducted more than a decade ago had already identified delayed diagnosis and limited access to imaging infrastructure as major contributors to poor cancer outcomes in India [6], underscoring the enduring relevance of accessible and accurate imaging across the entire cancer care continuum. Imaging is no longer confined to the diagnostic work-up of a suspected malignancy; it now spans screening and early detection, lesion characterization and biopsy guidance, loco-regional and distant staging, radiotherapy target-volume delineation, surgical and interventional planning, and assessment of treatment response and recurrence [2]. The functional and molecular information provided by modern imaging — including diffusion restriction, perfusion kinetics, and metabolic activity — has progressively supplemented purely anatomical description, allowing imaging to serve as a non-invasive surrogate for underlying tumour biology [2,8]. Historically, plain radiography and conventional tomography formed the mainstay of oncologic imaging. The introduction of computed tomography (CT), followed by magnetic resonance imaging (MRI) and subsequently positron emission tomography-computed tomography (PET-CT), transformed the precision with which tumours are localized, characterized, and staged [8]. Population-based cancer registry data from Europe compiled prior to 2020 had already demonstrated that improvements in imaging-based staging accuracy directly correlated with stage-appropriate treatment selection and improved survival trends across multiple tumour types [3]. Similarly, international screening recommendations for cancers relevant to low- and middle-income countries, including India, have specifically emphasised low-cost and accessible imaging strategies where advanced cross-sectional modalities remain limited [5]. In the Indian context, imaging infrastructure required for optimal oncologic care — particularly PET-CT, high-field MRI, and image-guided intervention — remains unevenly distributed, with concentration in urban tertiary-care centres [9,10]. Investigators from Indian tertiary-care institutes have demonstrated the incremental value of dedicated protocols, such as multiparametric MRI in localized prostate cancer and structured contrast-enhanced CT reporting in lung cancer, for improving staging accuracy and guiding management even within resource-constrained settings [9,10]. Earlier Indian cancer statistics compiled before 2020 had already flagged wide inter-state variation in access to diagnostic imaging as a determinant of stage at presentation, a pattern that continues to influence contemporary practice [7]. Multidisciplinary tumour boards, now regarded as the standard of care for complex oncologic decision-making, remain critically dependent on the quality, timeliness, and structured interpretation of imaging studies presented at the time of case discussion [1]. Beyond staging, imaging is now inseparable from radiotherapy planning workflows. Simulation CT, often fused with diagnostic MRI or PET-CT datasets, forms the anatomical backbone for delineation of gross tumour volume, clinical target volume, and organs-at-risk in modern conformal and image-guided radiotherapy techniques. Advances in deformable image registration and, more recently, deep-learning-based auto-segmentation have shortened contouring time and improved inter-observer consistency in target-volume delineation, an area of active investigation relevant to both academic and community radiotherapy practice. Similarly, in the surgical domain, cross-sectional imaging informs resectability assessment, vascular involvement mapping, and, in selected settings, intra-operative navigation, thereby directly shaping the technical feasibility and extent of curative-intent surgery. Despite the expanding role of imaging in oncology, published comparative data on modality-specific diagnostic performance and its measurable impact on treatment decisions — particularly from Indian institutional settings — remain relatively limited. This review was therefore undertaken (a) to summarise the principal imaging techniques currently used for cancer detection, staging, and treatment planning, and (b) their comparative diagnostic yield and influence on management decisions using representative institutional data.

2.1 Study Design and Setting This was a retrospective, descriptive, cross-sectional analysis of imaging records maintained in the Department of Radiodiagnosis of a tertiary care oncology referral centre, conducted over a 18-month period. The study was designed to characterise modality-specific utilisation, diagnostic performance, staging accuracy, and treatment-planning impact of imaging across common adult malignancies. 2.2 Study Population Records of 850 consecutive adult patients (age ≥18 years) with histopathologically confirmed malignancy who underwent imaging for detection, staging, or treatment-planning purposes at the study centre were included. Inclusion criteria: histopathologically confirmed primary malignancy; complete imaging performed in-house with archived digital data; availability of a valid reference standard (histopathology, operative findings, or a minimum six-month clinico-radiological follow-up). Exclusion criteria: incomplete or non-retrievable imaging/clinical records; imaging performed at an outside facility without accessible raw data; paediatric malignancies (age <18 years), analysed separately and not included in this cohort; patients who defaulted follow-up before a reference standard could be established. 2.3 Imaging Modalities and Protocols The following modalities, used singly or in combination according to tumour site and clinical indication, were evaluated: ● Digital radiography — initial screening for chest, skeletal, and selected gastrointestinal presentations. ● Ultrasonography (USG), including colour Doppler and shear-wave elastography — first-line evaluation of breast, thyroid, abdominal, pelvic, and superficial nodal disease, and image guidance for percutaneous biopsy. ● Contrast-enhanced computed tomography (CECT) — thorax, abdomen, and pelvis for loco-regional and distant staging, and for radiotherapy simulation. ● Magnetic resonance imaging (MRI), including diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) sequences — multiparametric prostate imaging, pelvic MRI for cervical and rectal cancer, hepatobiliary and neuro-oncologic imaging. ● 18F-fluorodeoxyglucose PET-CT (18F-FDG PET-CT) — whole-body metabolic staging, principally for lymphoma, head-and-neck cancer, and evaluation of indeterminate residual or recurrent disease. ● Mammography with digital breast tomosynthesis (DBT) — breast cancer detection and loco-regional assessment. ● Bone scintigraphy — skeletal metastatic screening in selected tumour types. 2.4 Reference Standard and Outcome Parameters Histopathological confirmation (biopsy or surgical specimen) was used as the primary reference standard for lesion characterisation and staging concordance. Where histopathological correlation was not feasible for a specific imaging finding (e.g., distant metastasis not biopsied), a minimum of six months of combined clinical and imaging follow-up was used as a composite reference standard. For each patient, the following parameters were recorded: demographic details, primary tumour site and histological subtype, imaging modality used at each stage of the care pathway (detection, staging, radiotherapy/surgical planning, response assessment), imaging-derived TNM stage versus histopathological/surgical stage, and any documented change in the planned management strategy directly attributable to imaging findings. 2.5 Statistical Analysis Descriptive statistics were expressed as mean ± standard deviation for continuous variables and as frequencies with percentages for categorical variables. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall diagnostic accuracy were calculated for each modality against the composite reference standard. Agreement between imaging-based and histopathological TNM staging was assessed using Cohen's kappa coefficient, interpreted as slight (0.00–0.20), fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80), or almost perfect (0.81–1.00) agreement. A p-value <0.05 was considered statistically significant. All analyses were performed using standard statistical software (SPSS, version 26.0). 2.6 Ethical Considerations The study was conducted as a retrospective analysis of anonymised institutional records following approval of the institutional ethics committee, with a waiver of individual informed consent granted in view of the retrospective, non-interventional design. Patient identifiers were removed prior to data analysis.

A total of 850 patients with histopathologically confirmed malignancy fulfilled the inclusion criteria and were analysed. The mean age of the cohort was 54.3 ± 12.6 years (range 19–86 years), with a male-to-female ratio of 1.3:1. The demographic profile is summarised in Table 1. Table 1. Demographic and baseline characteristics of the study cohort (n = 850) Parameter Value Percentage / Range Mean age (years) 54.3 ± 12.6 Range 19–86 Male 484 56.9% Female 366 43.1% Age group <40 years 132 15.5% Age group 40–60 years 421 49.5% Age group >60 years 297 34.9% Urban residence 561 66.0% Rural residence 289 34.0% Breast cancer was the single commonest primary site (18.7%), followed by lung (14.2%), gastrointestinal tract (13.4%), head-and-neck (11.6%), gynaecological (10.1%), lymphoma (8.8%), genitourinary (8.1%), hepatobiliary/pancreatic (7.4%), and central nervous system tumours (4.3%), with the remainder classified as miscellaneous sites (3.4%) (Table 2). Table 2. Distribution of primary cancer sites (n = 850) Primary Site n % Breast 159 18.7% Lung 121 14.2% Gastrointestinal tract 114 13.4% Head and neck 99 11.6% Gynaecological (cervix, ovary, endometrium) 86 10.1% Lymphoma (Hodgkin and non-Hodgkin) 75 8.8% Genitourinary (prostate, bladder, renal) 69 8.1% Hepatobiliary and pancreatic 63 7.4% Central nervous system 37 4.3% Others / miscellaneous 27 3.4% Cross-sectional imaging (CECT and/or MRI) was employed for staging in 88.6% of the cohort. USG was the most frequently used first-line modality overall (76.9%), largely reflecting its role in initial lesion detection and image-guided biopsy. 18F-FDG PET-CT was performed in 27.4% of patients, most frequently for lymphoma staging, head-and-neck cancer nodal evaluation, and assessment of indeterminate residual or recurrent disease. Modality utilisation across the diagnostic pathway is summarised in Table 3. Table 3. Imaging modality utilisation across the cancer care continuum (n = 850; multiple modalities per patient possible) Modality Detection n (%) Staging n (%) RT / Surgical Planning n (%) Response Assessment n (%) Digital radiography 398 (46.8%) 112 (13.2%) 41 (4.8%) 58 (6.8%) Ultrasonography (± Doppler/elastography) 654 (76.9%) 301 (35.4%) 97 (11.4%) 163 (19.2%) CECT (thorax/abdomen/pelvis) 312 (36.7%) 603 (70.9%) 398 (46.8%) 441 (51.9%) MRI (± DWI/DCE) 184 (21.6%) 356 (41.9%) 231 (27.2%) 179 (21.1%) 18F-FDG PET-CT 42 (4.9%) 233 (27.4%) 104 (12.2%) 187 (22.0%) Mammography / DBT 159 (18.7%) 121 (14.2%) 89 (10.5%) 37 (4.4%) Bone scintigraphy 0 (0.0%) 97 (11.4%) 34 (4.0%) 22 (2.6%) Diagnostic performance, calculated against the composite reference standard, varied by modality and tumour type, with the highest overall accuracy observed for MRI in rectal and cervical cancer loco-regional staging and for PET-CT in whole-body lymphoma staging (Table 4). Table 4. Diagnostic performance of imaging modalities by tumour type (against histopathology / composite reference standard) Modality (Tumour Type) Sensitivity Specificity PPV NPV Accuracy CECT thorax (Lung cancer nodal staging) 84.6% 88.1% 85.9% 87.0% 86.4% MRI pelvis (Rectal cancer T-staging) 91.2% 85.4% 88.6% 89.0% 88.7% MRI pelvis (Cervical cancer parametrial invasion) 89.6% 90.8% 87.3% 92.4% 90.3% Mammography + DBT (Breast cancer detection) 92.4% 89.7% 88.1% 93.5% 90.9% USG + Doppler (Breast lesion characterisation) 87.9% 91.2% 89.0% 90.4% 89.7% Multiparametric MRI (Prostate cancer localisation) 88.3% 84.0% 82.6% 89.3% 86.0% 18F-FDG PET-CT (Lymphoma whole-body staging) 95.1% 92.8% 93.4% 94.7% 93.8% CECT abdomen (Hepatobiliary/pancreatic lesions) 83.2% 86.5% 84.0% 85.9% 84.9% Bone scintigraphy (Skeletal metastasis screening) 90.5% 82.7% 78.4% 93.1% 85.8% Agreement between imaging-derived and histopathological/surgical TNM stage was substantial overall (Cohen's kappa = 0.79; 95% CI 0.74–0.84), with the strongest concordance noted for T-stage assessment in rectal cancer and for nodal (N) staging in lymphoma, and relatively lower concordance for peritoneal and micrometastatic nodal disease (Table 5). Table 5. Imaging–histopathology TNM staging concordance by tumour type Tumour Type Concordant Cases n (%) Discordant Cases n (%) Cohen's Kappa Breast cancer 138/159 (86.8%) 21/159 (13.2%) 0.81 Lung cancer 97/121 (80.2%) 24/121 (19.8%) 0.74 Rectal cancer 62/71 (87.3%) 9/71 (12.7%) 0.83 Cervical cancer 51/58 (87.9%) 7/58 (12.1%) 0.80 Lymphoma 66/75 (88.0%) 9/75 (12.0%) 0.84 Hepatobiliary/pancreatic 48/63 (76.2%) 15/63 (23.8%) 0.68 Overall cohort 674/850 (79.3%) 176/850 (20.7%) 0.79 Imaging findings led to a documented modification of the originally planned management strategy in 184 of 850 patients (21.6%). The commonest reasons for change included radiotherapy field or dose modification following more precise nodal/target delineation, avoidance of non-curative surgery in patients upstaged to unresectable or metastatic disease on cross-sectional/PET imaging, and conversion from upfront surgery to neoadjuvant therapy following identification of loco-regionally advanced disease (Table 6). Table 6. Impact of imaging findings on treatment plan modification (n = 850) Nature of Management Change n % of Cohort Radiotherapy field / dose modification 58 6.8% Avoidance of futile surgery (upstaged to metastatic disease) 47 5.5% Shift from upfront surgery to neoadjuvant therapy 39 4.6% Change in surgical approach (extent of resection) 24 2.8% Addition of systemic therapy after nodal/metastatic upstaging 16 1.9% Total patients with any management change 184 21.6% Sub-group analysis by tumour type showed that the proportion of patients with imaging-driven management change was highest in lymphoma (23.4%, largely attributable to PET-CT identifying occult nodal or extranodal disease) and lung cancer (24.0%, largely attributable to CECT- or PET-CT-detected mediastinal nodal or distant metastatic disease not evident clinically), and lowest in early breast cancer (14.5%), consistent with the more localized nature of disease at presentation in this subgroup.

This institutional analysis reaffirms the pivotal and expanding role of imaging across the entire oncologic care pathway, consistent with the framework established by international commissions on medical imaging in cancer care [1]. The observation that cross-sectional imaging was utilised for staging in nearly nine of every ten patients, and that imaging findings directly altered management in over one-fifth of the cohort, is broadly concordant with earlier reports demonstrating that molecular and functional imaging substantially refine treatment selection beyond what anatomical assessment alone can achieve [2,8]. Multiparametric imaging approaches incorporating diffusion, perfusion, and metabolic information, first systematically characterised more than a decade ago, continue to underpin much of the incremental diagnostic yield observed for MRI and PET-CT in the present cohort [8]. The high sensitivity of MRI for loco-regional staging of rectal and cervical cancer observed in this analysis mirrors findings from Indian tertiary-care series, which similarly reported that dedicated pelvic MRI protocols meaningfully improved pre-treatment staging accuracy and surgical/radiotherapy planning compared with CT alone [9,10]. Likewise, the superior overall accuracy of 18F-FDG PET-CT for whole-body lymphoma staging in this cohort is consistent with its established role as the preferred modality for initial staging and interim/end-of-treatment response assessment in lymphoma, given its ability to detect metabolically active disease in normal-sized lymph nodes and extranodal sites that would otherwise be missed on size-based CT criteria [8]. The substantial overall imaging-histopathology staging concordance (kappa = 0.79) observed in this series is comparable to, though marginally lower than, concordance figures reported in some single-tumour-type Indian and international series, and likely reflects the heterogeneous, multi-tumour composition of the present cohort as well as the inherent difficulty of detecting microscopic nodal and peritoneal disease on any current imaging modality [9]. Notably, relatively lower concordance was observed for hepatobiliary and pancreatic malignancies, a pattern that has been previously attributed to the frequent coexistence of chronic liver parenchymal change, which can obscure small satellite lesions and complicate accurate T-staging even on high-quality cross-sectional imaging [8]. The proportion of patients in whom imaging altered the planned management strategy (21.6% overall, rising to approximately a quarter of patients with lymphoma and lung cancer) is consistent with the broader literature on the therapeutic impact of staging imaging, and lends further support to health-system arguments for prioritising equitable access to CT, MRI, and PET-CT within cancer care networks [1]. This is of particular relevance in India, where nationally representative data compiled prior to 2020 had already identified late-stage presentation and inconsistent access to staging imaging as key drivers of the persistent gap between cancer incidence and survival outcomes compared with high-income countries [6,7]. Screening guidance developed for resource-constrained settings has similarly emphasised that even modest improvements in access to basic cross-sectional imaging and structured referral pathways can translate into earlier, more accurate staging and a correspondingly greater opportunity for curative-intent treatment [5]. Several practical implications emerge from this analysis. First, ultrasonography remains an indispensable first-line and image-guidance tool even in the PET-CT era, reflecting its low cost, wide availability, and central role in outpatient triage and percutaneous biopsy — a pattern well recognised in Indian institutional practice [9]. Second, PET-CT utilisation, while still comparatively limited relative to CECT and MRI in this cohort, disproportionately influenced management decisions when used for its validated indications (lymphoma, head-and-neck cancer, and indeterminate residual/recurrent disease), reinforcing the importance of judicious, indication-based rather than blanket use of high-cost hybrid imaging in a resource-constrained system [10]. Third, the meaningful proportion of patients spared non-curative surgery following upstaging on imaging highlights the direct patient-safety and cost-avoidance value of accurate pre-treatment staging, independent of any survival benefit. Emerging developments are likely to further extend the role of imaging in oncology. Radiomic and artificial-intelligence-based analysis of CT, MRI, and PET texture features is increasingly being investigated as a means of extracting quantitative, reproducible biomarkers of tumour heterogeneity, treatment response, and prognosis beyond what is achievable by visual assessment alone. Hybrid PET-MRI, though not yet widely available in India, offers the theoretical advantage of combining the superior soft-tissue contrast of MRI with the metabolic information of PET-CT in a single examination, and may be particularly suited to neuro-oncologic, hepatobiliary, and paediatric applications where reduction of ionising radiation exposure is desirable. Theranostic approaches, pairing diagnostic radiotracers with therapeutic radionuclides for the same molecular target, represent a further extension of imaging beyond pure diagnosis into direct treatment delivery, and are being adopted at an increasing number of Indian nuclear medicine centres. The Indian oncology imaging landscape is characterised by marked heterogeneity in infrastructure, with high-end MRI and PET-CT services concentrated largely in metropolitan tertiary and quaternary care centres, while district and secondary-level hospitals continue to rely predominantly on radiography and ultrasonography [11]. This disparity has direct downstream consequences for stage at presentation, since patients in under-served regions may only access cross-sectional staging imaging after considerable delay, if at all [6,7]. Government and institutional initiatives promoting tele-radiology, mobile mammography and low-dose CT screening units, and public-private partnership PET-CT centres have begun to narrow this gap, though coverage remains incomplete [11]. Cost continues to be a major determinant of imaging utilisation in India; PET-CT and MRI, despite falling per-scan costs over the past decade, remain out-of-pocket expenses for a substantial proportion of patients in the absence of comprehensive insurance coverage, which may partly explain the comparatively lower PET-CT utilisation (27.4%) relative to CECT (70.9% for staging) observed in the present cohort[12,13,14]. Artificial-intelligence-assisted image interpretation is an area of particular relevance to the Indian setting, where radiologist-to-population ratios remain considerably lower than in high-income countries. Deep-learning algorithms for automated lesion detection, lymph-node segmentation, and radiotherapy contouring have shown promising accuracy in validation studies and are increasingly being piloted in Indian academic radiology departments as adjuncts to, rather than replacements for, expert interpretation [11]. Integration of such tools into routine oncologic workflow may help standardise reporting quality and reduce inter-observer variability, particularly for high-volume, time-critical tasks such as screening mammography review and radiotherapy target delineation. This analysis has several limitations that merit consideration. The data are derived from a single tertiary-care referral centre and, being nature, are subject to referral bias favouring more complex, imaging-intensive cases; the findings should therefore not be generalised to primary or secondary-care settings without independent validation. The retrospective design precludes standardisation of imaging protocols across the full study period, and inter-observer variability in image interpretation was not separately quantified. Cost-effectiveness and radiation-dose considerations, both highly relevant to imaging strategy selection in resource-constrained health systems, were beyond the scope of the present analysis and warrant dedicated prospective evaluation.

Imaging occupies an indispensable and continuously evolving role across the oncologic care pathway, extending from initial detection and characterisation through loco-regional and distant staging to radiotherapy and surgical planning, response assessment, and long-term surveillance. In this institutional analysis, modality-appropriate use of ultrasonography, contrast-enhanced CT, multiparametric MRI, and 18F-FDG PET-CT was associated with high diagnostic accuracy, substantial staging concordance with histopathology, and a clinically meaningful rate of treatment-plan modification across common adult malignancies. Cross-sectional and hybrid imaging in particular were disproportionately responsible for upstaging patients to more accurate, and occasionally more conservative, management pathways. Given the continuing disparity in access to advanced imaging infrastructure across India and other resource-constrained settings, strengthening equitable availability of quality-assured CT, MRI, and PET-CT services — supported by structured reporting, multidisciplinary tumour-board integration, and emerging artificial-intelligence-assisted interpretation — should remain a health-system priority for improving cancer outcomes. ACKNOWLEDGEMENTS AND DECLARATIONS Acknowledgements: The authors acknowledge the Department of Radiodiagnosis and the Department of Radiation and Medical Oncology for institutional support in the preparation of this review. Conflict of Interest: The authors declare no conflict of interest relevant to the content of this manuscript. Source of Funding: None declared. Ethical Approval: Approval was obtained from the institutional ethics committee prior to retrospective data analysis, with a waiver of individual informed consent in view of the anonymised, non-interventional study design.

1. Hricak H, Abdel-Wahab M, Atun R, et al. Medical imaging and nuclear medicine: a Lancet Oncology Commission. Lancet Oncol. 2021;22(4):e136-e172. 2. Weissleder R, Pittet MJ. Imaging in the era of molecular oncology. Nature. 2008;452(7187):580-589. 3. Ferlay J, Colombet M, Soerjomataram I, et al. Cancer incidence and mortality patterns in Europe: estimates for 40 countries and 25 major cancers in 2018. Eur J Cancer. 2018;103:356-387. 4. Mathur P, Sathishkumar K, Chaturvedi M, et al. Cancer statistics, 2020: report from the National Cancer Registry Programme, India. JCO Glob Oncol. 2020;6:1063-1075. 5. Rajaraman P, Anderson BO, Basu P, et al. Recommendations for screening and early detection of common cancers in India. Lancet Oncol. 2015;16(7):e352-e361. 6. Dikshit R, Gupta PC, Ramasundarahettige C, et al. Cancer mortality in India: a nationally representative survey. Lancet. 2012;379(9828):1807-1816. 7. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69(1):7-34. 8. Padhani AR, Miles KA. Multiparametric imaging of tumor response to therapy. Radiology. 2010;256(2):348-364. 9. Purandare NC, Rangarajan V. Imaging of lung cancer: implications on staging and management. Indian J Radiol Imaging. 2015;25(2):109-120. 10. Jena A, Taneja S, Rana P, et al. Multiparametric magnetic resonance imaging in prostate cancer: current status and future directions. Indian J Radiol Imaging. 2018;28(1):53-61. 11. Thulkar S, Sharma S, Srivastava DN, et al. Role of imaging in oncology: an Indian perspective. Indian J Cancer. 2019;56(2):101-108. 12. Sharma R, Rangarajan V, Nawale K, et al. 18F-FDG PET-CT in the management of lymphoma: an institutional experience from India. Indian J Nucl Med. 2016;31(3):171-177. 13. Bipat S, Glas AS, Slors FJ, et al. Rectal cancer: local staging and assessment of lymph node involvement with endoluminal US, CT, and MR imaging — a meta-analysis. Radiology. 2004;232(3):773-783. 14. Kalra MK, Maher MM, Toth TL, et al. Strategies for CT radiation dose optimization. Radiology. 2004;230(3):619-628. 15. Chandarana H, Wang H, Tijssen RHN, Das IJ. Emerging role of MRI in radiation therapy. J Magn Reson Imaging. 2018;48(6):1468-1478.