In this study, we compared postmortem photon-counting computed tomography (PMPCCT) with conventional postmortem computed tomography (PMCT), evaluating both objective and subjective image quality across several reconstruction kernels and slice thicknesses. Image quality was considered in terms of objective metrics (SNR, CNR, noise) as well as subjective ratings (sharpness, bone detail, contour visibility, artifacts), reflecting the different diagnostic demands of soft tissue (contrast-based) and bone (resolution-based) imaging. Findings under our scanning conditions suggest that PMPCCT offers advantages at moderate slice thicknesses but underperformed in specific scenarios, particularly at ultra-thin slices and in lung imaging. However, its performance is influenced by slice thickness, reconstruction algorithms, and tissue type.

These results are significant as they highlight PMPCCT’s potential to improve postmortem imaging, an area where PMCT often faces challenges. Additionally, the study emphasizes the complexity of postmortem imaging, noting that phantom-based studies do not fully replicate the intricacies of real-world decedent imaging. This underscores the need for further research to optimize PMPCCT for forensic applications. Understanding the nuanced performance of PMPCCT under varying technical parameters can aid in selecting and tailoring imaging protocols for postmortem examinations.

In general, the results show that PMPCCT performed best at slice thicknesses of 0.4 mm, while its advantages diminished at 0.2 mm, possibly due to reduced photon statistics and increased noise [7, 24]. This effect is consistent with some of the previous findings on diminishing returns at extreme resolutions [23, 29].

Aditionally, reconstruction kernels appeared to play a role in image quality differences. The softer Br40 kernel likely helped PMPCCT by reducing noise, whereas the sharper Br60 kernel may have amplified noise, favoring PMCT [6, 28]. This could have been accentuated due to differences in reconstruction algorithms, as PMCT’s ADMIRE 3 is optimized for strong noise suppression [3, 5]. However, the inferior performance of PMPCCT at Br60 in decedent imaging was rather unexpected, as literature generally supports PCCT’s superiority even with sharper kernels [6, 16, 29].

In lung imaging, PMCT outperformed PMPCCT in our study, particularly at thinner slice settings. This result agrees with earlier research showing that PCCT can be prone to elevated high-frequency noise in aerated tissues [20, 27]. Conversely, PMPCCT demonstrated advantages in soft tissue and liver imaging, particularly under Br40 conditions. This is in line with existing literature describing PCCT’s superior soft tissue contrast and resolution [10, 25, 31, 33].

Reconstruction parameters represent an important technical factor in diagnostic assessment. Matrix size directly affects spatial resolution: a 1024 × 1024 matrix, as available on PMPCCT, increases pixel density and improves detail visibility, particularly for bone structures, but also enhances the perception of noise. Slice thickness has a similar bidirectional effect: thinner slices reduce partial volume averaging and improve sharpness but simultaneously increase noise, thereby lowering SNR and CNR. In this study, PMPCCT reconstructions included thinner slices and a larger matrix compared to PMCT, reflecting the standard defaults of each system. These settings may have contributed to the observed improvements in sharpness and bone detail and must therefore be considered potential confounders when interpreting the results.

Additional factors may have influenced performance. Both PMCT and PMPCCT were performed with relatively high tube voltages, as radiation dose is not a limiting factor in postmortem imaging. In clinical settings, PCCT is considered particularly advantageous because it delivers optimal image quality even at lower radiation doses [26, 34]. However, since this benefit is irrelevant in postmortem imaging, the generally higher tube voltages used may have mitigated some of PMPCCT’s expected advantages—an aspect that may warrant further investigation [22, 35]. In this context, it is also relevant that – although scan protocols and scanning conditions were kept as similar as possible between PMPCCT and PMCT – slightly higher CTDI values were ultimately reached in the PMPCCT scans of the decedent. As a result, the associated improved outcomes should be interpreted with caution.

A notable discrepancy emerged between phantom and decedent imaging. Since phantoms consist of homogeneous materials, they minimize scattering effects, whereas decedent scans involve heterogeneous structures that likely led to more complex signal distributions [30, 36]. Prior studies have reported similar deviations between phantom-based and real world imaging outcomes [12, 36], reinforcing the need for caution when extrapolating results. Furthermore, most PCCT studies focus on clinical rather than postmortem imaging so far. Unique postmortem factors, such as the absence of motion artifacts and altered tissue properties, dose independency and body temperature, may influence the comparative performance of PMPCCT and PMCT [30, 37].

The subjective image quality assessment of PMPCCT should be interpreted as indicative rather than conclusive. Due to the small sample size statistical analysis was not feasible. Overall, PMPCCT demonstrated subjective advantages in image quality, sharpness, and bone details while performing similarly to PMCT in contour visibility, noise, and artifact reduction. These findings align with existing literature on PCCT in clinical settings, which consistently highlights improved spatial resolution and contrast [7, 10, 15,16,17,18, 30, 35, 36].

The comparable noise perception between PMPCCT and PMCT is consistent with prior research, as overall noise levels depend on reconstruction algorithms and dose settings [28, 29].

The similar performance in artifact formation, especially in the presence of an ICD catheter, is notable. PCCT is suggested to reduce metal artifacts due to its energy-resolving capabilities [11, 15, 17, 36]; however, its effectiveness varies depending on the metal composition and reconstruction techniques used [6, 13, 14, 31]. The similar performance in artifact formation in the presence of an ICD catheter in our study is not very surprising, since no specialized artifact-reducing programs were applied. Follow-up studies would be necessary to objectively assess any potential advantage in this regard.

A major limitation of this study is its small sample size, including only one decedent and one phantom, and its controlled experimental conditions. This prevents generalization of the results and precludes any meaningful statistical testing. The study should therefore be regarded as an exploratory pilot investigation, providing first insights into the feasibility and potential of PMPCCT in postmortem imaging. While the findings provide valuable insights, larger studies involving diverse cases and real-world postmortem scenarios are essential to validate the results and account for the variations in tissue composition and imaging challenges in actual forensic investigations.

The discrepancies between phantom and decedent imaging highlight the need for studies conducted under real-world postmortem conditions. Factors such as body temperature, degree of decomposition, and cause of death are likely to influence imaging outcomes and cannot be replicated in phantoms. Although the BMI of the included decedent was within the normal range, differences in body habitus may also affect image quality through variable tissue attenuation. To improve generalizability, future investigations should therefore include cases across a broader BMI spectrum and encompass a wider range of postmortem changes. Such studies will be essential to assess PMPCCT performance under diverse conditions and to optimize imaging protocols for forensic application.

Even though the scanning conditions were kept as consistent as possible, differences in reconstruction parameters between PMCT and PMPCCT, including slice thickness (0.5 mm for PMCT vs. 0.4 mm and 0.2 mm for PMPCCT), reconstruction algorithms, and slightly varying dose metrics, may have influenced image quality, SNR, CNR, and subjective ratings independent of detector technology [2, 6, 7, 16, 20, 23, 28, 29]. These settings were chosen according to standard usage of each system to reflect routine practice and to achieve optimal image quality. Nevertheless, such differences must be considered a potential source of bias. Future studies should therefore investigate harmonized and optimized PCCT reconstruction protocols tailored for postmortem imaging, especially at ultra-thin slices and in complex anatomical regions [6, 16, 28].

ROI measurements were obtained across five consecutive slices per anatomical region by one reader and verified by a second. While this approach improves robustness, reproducibility was not formally assessed and inter- and intraobserver variability remain potential limitations.

Finally, our study was conducted using a 140 kV tube voltage, which may have diminished some of the expected advantages of PMPCCT. Since radiation dose is not a limiting factor in postmortem imaging, future research should explore the effects of lower kV settings to fully exploit the benefits of photon-counting technology, as suggested by clinical applications [22, 26].

Furthermore, it will be essential to investigate the spectral imaging capabilities unique to PMPCCT to differentiate between various tissue types and foreign bodies, which could be especially useful in forensic imaging [6, 10, 11, 13,14,15, 17, 18, 32, 38].

Optimizing scan protocols and reconstruction algorithms is a critical step toward enhancing PMPCCT’s performance. Such advancements will not only refine the imaging process but also strengthen the diagnostic potential of PCCT in forensic applications.