Total body PET/CT: Future aspects

Abstract

Total-body (TB) positron emission tomography (PET) scanners are classified by their axial field of view (FOV). Long axial field of view (LAFOV) PET scanners can capture images from eyes to thighs in a one-bed position, covering all major organs with an axial FOV of about 100 cm. However, they often miss essential areas like distal lower extremities, limiting their use beyond oncology.TB-PET is reserved for scanners with a FOV of 180 cm or longer, allowing coverage of most of the body. LAFOV PET technology emerged about 40 years ago but gained traction recently due to advancements in data acquisition and cost. Early research highlighted its benefits, leading to the first FDA-cleared TB-PET/CT device in 2019 at UC Davis. Since then, various LAFOV scanners with enhanced capabilities have been developed, improving image quality, reducing acquisition times, and allowing for dynamic imaging. The uEXPLORER, the first LAFOV scanner, has a 194 cm active PET AFOV, far exceeding traditional scanners. The Panorama GS and others have followed suit in optimizing FOVs. Despite slow adoption due to the COVID pandemic and costs, over 50 LAFOV scanners are now in use globally.

This review explores the future of LAFOV technology based on recent literature and experiences, covering its clinical applications, implications for radiation oncology, challenges in managing PET data, and expectations for technological advancements.

1. Introduction

Total-body (TB) positron emission tomography (PET) scanners can be clinically classified according to their axial field of view (FOV) [1]. Long axial field of view (LAFOV) PET is any scanner that can scan from eyes to thighs in a one-bed position, defined by an axial FOV (AFOV) of approximately 100cm, where all major organs are covered. However, these scanners often lack coverage of essential body parts, such as distal lower extremities, for imaging beyond oncologic indications, e.g., musculoskeletal (MSK) or inflammation diagnostics. Therefore, the term TB-PET is restricted to scanners with AFOV lengths of 180cm or longer that can scan most of the population’s entire body in a one-bed position [2]. LAFOV PET scanners were conceived initially over 40 years ago. However, they have recently become a reality for many reasons, including cost and adequate data acquisition technology availability. The theoretical benefits of LAFOV PET were explored by different research groups in several proposals and simulation studies, particularly to study the benefits of the signal collection efficiency gain [3], [4], [5], [6], [7], [8], [9]. These early studies paved the way for the first commercially available FDA-cleared TB-PET/CT device in the US at the University of California, Davis (UC Davis) in 2019 [10]. Since then, several other LAFOV PET scanners with different axial lengths have been developed [11], [12], [13], [14], [15], [16], [17], [18], [19]. Their improved signal collection efficiency and state-of-the-art high spatial resolution have opened the door to enhanced clinical applications previously unavailable with standard FOV commercial scanners. These imaging enhancements include (a) reduced image acquisition times, (b) reduced administered activity, (c) the capability of acquiring scans at later time points using the same initial dose, (d) improvement in image quality, and (e) the capability of capturing dynamic processes over the entire body.

The first generation LAFOV scanner was the uEXPLORER, developed through a collaboration between UC Davis and United Imaging Healthcare (UIH), funded by a Transformative R01 award from the National Institutes of Health. The scanner is composed of a total-body PET with a CT tomograph; it has a transaxial FOV of 68.6 cm and an active PET AFOV of 194.0cm [11], much longer than conventional PET scanners with AFOVs varying from approximately 15 cm to 35 cm [20]. It is worth noting that the second-generation scanner, the Panorama GS, has a 148cm AFOV[21].

In the following five years, this type of scanner spread worldwide. For instance, the PennPET EXPLORER built at the University of Pennsylvania originally had an AFOV of 64cm [18], which has now been updated to 142cm, and the Siemens Biograph Vision Quadra with a 106cm AFOV, installed at Inselspital University Hospital in Bern [22], [23], were the scanners designed to investigate and exploit the LAFOV benefits. The spread of LAFOV scanners has been relatively slow; possible reasons include the COVID pandemic and costs. Nevertheless, this technology is now widely disseminated with over 50 scanners in Oceania, the Americas, Europe, and Asia. The LAFOV technology has advanced to the point where we can now begin to explore its clinical capabilities. This invites the question, what does the future of LAFOV development look like? This review provides future perspectives on LAFOV technology based on an analysis of the most recent published literature and our experiences. The review will cover the future clinical use of the technology. We will present the impact of LAFOV PET-based dosimetry on radiation oncology. We will discuss the future challenges of reporting vast amounts of PET data and the effects of potentially identifying patients directly from high-quality (TB) PET images. Finally, we will discuss the future expectations for the design and technological development of LAFOV PET.

2. LAFOV PET scanner design overview

Comparing LAFOV scanners like uEXPLORER, Panorama GS, and Siemens Biograph Vision Quadra opens the “optimal” FOV length discussion. This section will discuss the driving rationale and design factors for LAFOV PET scanners. Table 1 overviews the design parameters investigated for long-axial FOV PET scanners.

Table 1.

Relevant scanner parameters and metrics used for quantiative comparison. References: (a)[11], (b)[23], (c)[12] , (d) [24], (e)[13], (f)[14], (g)[25], [26], [27], (h)[16], (i)[17], (j)[18]

Existing Builds
Scanner	Spatial Resolution at axial center*	TOF resolution	NEMA NU-2 sensitivity	Peak NECR	Length AFOV	Crystal size
uEXPLORER	~ 2.9 mm(a)	505 ps(a)	174 kcps/MBq(a)	1524 kcps(a)	194 cm(a)	2.76 × 2.76 × 18.1 mm3 (a)
Siemens Biograph Vision Quadra	~ 3.5 mm(b)	228 ps(b)	176 kcps/MBq(b)	2956 kcps(b)	106 cm(b)	3.2 × 3.2 × 20.0 mm3 (b)
uMI Panorama GS	~ 2.9 mm(d)	188 ps(c)	176 kcps/MBq(c)	3350 kcps(c)	148 cm(c)	2.76 × 2.76 × 18.1 mm3 (c)
PennPET Explorer	~ 4.0 mm(f)	250 ps(f)	140.2 kcps/MBq(f)	1550 kcps(f)	142 cm(f)	3.86 × 3.86 × 19.0 mm3 (j)
Future builds
J-PET **	~ 5.8 mm without DOI(e)	~ 220 ps(e)	38 kcps/MBq(e)	630 kcps(e)	200 cm(e)	2000 × 6 × 30 mm3 (e)
Walk through PET **	~ 2.0 mm(g)	327 ps(g)	154 kcps/MBq(g)	2.61 kcps(g)	106 cm(g)	16 × 50 × 50 mm3 (i)

^*The spatial resolution represents the average of the radial, tangential and axial components.

^**The data for J-PET originate from simulations

The initial motivating factor for the LAFOV PET scanner design was the detection sensitivity and the coverage for multi-organ imaging, which depends highly on axial length. Increasing the axial length, although it is not trivial, it can be as simple as increasing the number of detector rings. However, because time-of-flight (TOF) can also improve the quality of the detected counts independent of geometrical sensitivity, it has become a research focus in both industry and academia. TOF resolution is the precision with which the time difference between the arrival of the two coincident photons from a positron decay can be measured, it may also be called the system’s coincidence timing resolution (CTR). TOF is influenced by several factors, including the type of scintillator, the coupling to the photon detector, the detector type, and the read-out electronics [28], [29]. Improved TOF resolution enhances the image signal-to-noise ratio (SNR)[29] by reducing the number of random events and the uncertainty in the position of the detected true events along the relevant line of response, thus improving the results of the image reconstruction process. Since SNR is proportional to the square root of the number of detected events [30], a better TOF resolution can be said to increase the system’s effective sensitivity [33]. Improved TOF resolution accelerates the convergence of iterative image reconstruction methods, such as maximum likelihood or penalized likelihood reconstructions. Fewer iterations are associated with reduced image noise, meaning that enhancing TOF resolution positively impacts the trade-off between noise level, scan time, and lesion contrast. The “ideal” TOF resolution would allow for the direct determination of the source location with millimeter-level precision, enabling the generation of cross-sectional images directly from the measured TOF PET data. Achieving “reconstruction-free” PET imaging would require a CTR on the order of a few tens of picoseconds. This formidable challenge has prompted significant efforts to create a “roadmap” for reaching this goal [31], and progress has been made, with at least one benchtop system demonstrating reconstruction-free PET [32]. However, no commercial scanner or practical prototype currently exists that can do this. So, future PET scanners may not need to have an LAFOV to achieve a high-sensitivity performance.

Another critical metric to consider in designing a PET scanner is the spatial resolution, which historically has been poor compared to other modalities. In PET scanners, a key factor in spatial resolution is the dimensions of the crystals that compose the PET detector crystal array, both the axial and transaxial directions. For instance, the uEXPLORER, with a crystal side length of 2.76 mm, achieves a spatial resolution of approximately 3 mm. In comparison, the Biograph Vision Quadra has a resolution of 3.5 mm, constrained by its crystal side length of 3.2 mm [35].

Cost of construction plays an important role in the advancement of PET technology. The crystals in the PET detector tend to be the most expensive item. Hence the development of low-cost systems to increase the user base will influence the proliferation and evolution of LAFOV PET scanners. The J-PET system (table 1) is the first low-cost LAFOV PET scanner, utilizing affordable scintillation crystals made from polymeric materials.

3. Clinical aspects of TB PET/CT

The advent of TB-PET/CT scanners has marked a significant milestone in molecular imaging, revolutionizing the clinical capabilities within the field [33], and it has become the subject of various clinical research studies[2], [20], [34] . LAFOV scanners have some imaging advantages, such as increased signal collection efficiency and high spatial resolution compared to conventional short-axial field-of-view (SAFOV) PET systems[22], [35]. This has made detecting small lesions, such as coronary artery plaques, possible with LAFOV PET [36]. Moreover, due to LAFOV PET’s higher sensitivity, reducing injected activities is also feasible [37].

Low-dose imaging is especially of clinical interest in vulnerable patient cohorts like pregnant women or children [38], [39]. Recently, it has been demonstrated that [¹⁸F]FDG TB-PET/CT with 1/8^th of the clinical reference standard dose is feasible in a cohort of 18 pediatric patients [40]. A short acquisition time is also feasible with TB and LAFOV PET/CT. With short acquisition protocols, PET centers can not only increase their throughput [29], but these protocols can also reduce motion artifacts and increase the comfort of the patients [41].

3.1. Organ-specific connection with dynamic TB-PET/CT

LAFOV systems offer unique possibilities when studying different organs and their relationships. [42]. For example, LAFOV allows endocrinologic studies that investigate the organ-to-organ interactions involved when stress is present, such as studying the hypothalamic–pituitary–adrenal axis, which regulates mood, digestion, immune function, and energy storage. It is now possible to perform many other “whole-person” studies [43]. Currently, there is no clear understanding of the intraorgan relationships expected; however, it is anticipated that TB-PET will uncover causal relationships between organs, potentially leading to significant medical interventions.

Simultaneous dynamic imaging allows the analysis of molecular processes, characterizing parameters such as the net molecular influx or perfusion. For instance, Wang et al. showed that FDG can provide the tissue uptake rate, K1, and the net influx rate Ki maps, which are surrogate for perfusion and net influx, respectively, across the entire body[44]e challenges may occur when applying parametric imaging protocols on LAFOV and TB PET/CT systems. Parametric imaging, for example, may require long acquisition times, possibly exceeding 60 minutes. This restriction requires adequate patient selection prior to the scan and non-painful patient positioning to complete these protocols successfully. In addition, various approaches have focused on shortening dynamic scan protocols[45]. Li et al. proposed a self-supervised deep-learning method to reduce the noise in parametric images with short dynamic scans[46]. Viswanath et al. proposed a protocol of dynamic 10–15 min scanning directly after the tracer injection followed by a 5 min scan 60 min post-injection on a LAFOV PET/CT scanner [47]. The missing dynamic data could be interpolated between the two time points.

Additionally, Sari et al. showed that one-time point kinetic modeling in oncological patients was feasible using a population-based input function with [¹⁸F]FDG LAFOV PET/CT [48]. Moreover, motion correction is warranted if scanning dynamically with LAFOV PET/CT due to patient movement over time. Artificial intelligence (AI)- aided tools may help to perform motion correction with TB and LAFOV PET/CT and are currently the subject of research efforts [49].

Practical dynamic imaging and multi-organ visualization can change the future of patient management in cases where diagnosis is difficult. Moreover, the use of AI for disease detection will be further developed, and performance will be improved with the bountiful data that LAFOV PET/CT scanners can produce. With much more data available, LAFOV PET systems are advancing towards a single scan diagnosis approach across many physiological systems.

4. Development of new tracers and therapies

The enhanced capabilities of TB PET and LAFOV PET systems [40], [50] enable more comprehensive evaluations of the pharmacokinetic and pharmacodynamic characteristics of new drug candidates. Imaging the entire body in a single acquisition allows improved accuracy and flexibility in multi-tracer imaging [51], accurate dosimetry of therapies that emit β+ particles [52], and allows for flexibility in patient management, as images taken later can be diagnostically valuable[53].

Dynamic imaging allows for the real-time capture of tracer distribution and kinetics across extensive anatomical regions, facilitating a comprehensive analysis of physiological processes, and it may even be possible to use it to measure perfusion with tracers that have a low first-pass extraction fraction, opening up a large area of possibilities for measuring complementary biomarkers[54]. In the case of infectious diseases, capturing total-body radiotracer activity with dynamic imaging allows for a better understanding of the immune cell distribution and trafficking inside all tissues in humans [55], [56].

The enhanced imaging capabilities of LAFOV PET may also enable simultaneous imaging with multiple tracers to improve the detection and characterization of abnormal tissue in areas such as oncology, cardiology, and neurology. For example, using a combination of 68Ga-PSMA-11 and 18F-DG PET/CT could help in selecting the right patients for new theragnostic treatments [57]. Multi-tracer imaging may offer precise detection and characterization, especially when the tracers have different effective half-lives, allowing for identification through dynamic or multi-time point imaging.

LAFOV PET enables the measurement of radiopharmaceutical therapy dose distribution with enhanced image resolution in the abdominal region. For instance, LAFOV PET shows superior accuracy in 90Y radioembolization dosimetry in part due to the capacity to image both lungs and liver with a single bed acquisition with increased sensitivity [52]. Furthermore, the high sensitivity of LAFOV PET establishes it as a prime candidate for predicting the efficacy of radiopharmaceutical therapies using time-activity curves. Although these methods have been explored using conventional PET systems, the poorer sensitivity of these systems may compromise accuracy[58], [59].

Finally, delayed imaging with clinically available radiotracers may improve cancer detection; several examples of this concept are reported in the literature for both PSMA-labelled, DOTATATE-labelled radiotracers and the widely used 18F-FDG molecule [53], [60]. The enhanced signal collection efficiency of LAVOF overcomes/mitigates the decrease in signal even when maintaining acquisition times similar to standard protocols.

The continuous advancements in LAFOV PET technology, particularly its integration with modalities such as perfusion-CT, promise to significantly advance the field of molecular imaging [61]. Non-contrast CT is already an integral component[62], [63], and the combination of perfusion CT with dynamic PET information could lead to an even more precise analysis of the differences between image uptake and blood supply. On the other hand novel FDG perfusion PET based on dynamic imaging have been demonstrated in the brain[54]. Either way, this enhanced capability is expected to improve human drug development processes and deepen our understanding of biological dynamics.

5. Technological and Hardware advances

For several decades, LAFOV and TB PET scanners have been proposed, researched, and developed [1], [64], [65], [66]. In this section, we will summarize recent technical developments in LAFOV and TB PET scanning and highlight the most pressing central research questions related to scanner length, detector optimization, and cost considerations in the context of future expectations.

When increasing the length of a PET scanner, the most immediate benefit is a gain in system sensitivity. Sensitivity, in simple terms, refers to the fraction of emitted coincident photons that the scanner can detect, typically measured through a line source experiment as per NEMA standards [67]. Enhanced sensitivity positively impacts image quality and enables shorter scan times, lower radiation doses, late time-point scanning, or a combination of these benefits [3]. However, these advantages come with trade-offs, including increased space requirements (at least, for TB-PET), higher procurement and maintenance costs, and greater computational demands for image reconstruction, data corrections, and data storage.

Time of flight performance:

As discussed in section 2, the performance of a scanner can also be improved with TOF. The goal is to know the precise location of the origin of the annihilation event. A breakthrough occurred in 2021 when Kwon et al. demonstrated that, in the absence of scattered coincidences, it is practically possible to image positron emitters directly without reconstruction using a coincident measurement with two detectors achieving a CTR of 32 ps [68]. Despite this advance, attenuation compensation and iterative image reconstruction techniques are still necessary. Theoretically, a 10ps TOF will enable PET imaging without needing reconstruction, achieving millimeter spatial resolution, presenting an exciting challenge for PET detector developers [74]. Even with a TOF resolution of 10ps, simple direct (reconstruction-free) methods might not yield the optimal image quality in terms of resolution and noise compared to iterative reconstruction techniques [69]. Therefore, further research into the image generation process will be required even if the technical hurdles to achieve 10 ps resolution are overcome.

The next-generation LAFOV PET scanners are expected to have significantly better TOF, expanding the use of high SNR PET images in many clinical applications.

Scintillator design:

Another crucial aspect of the design of LAFOV and TB PET scanners is the size and type of the scintillator crystals, which significantly impact scanner performance, including sensitivity, energy resolution, and TOF resolution [28]. For instance, Lutetium-based scintillators like LSO and LYSO are widely used in commercial and research scanners due to their high light yield (32 photons/keV) and short decay time (41 ns), which enable excellent energy resolution and timing capabilities[1]. While the best timing performance can be achieved with more affordable plastic scintillators such as BC-408 (2.1 ns), the long attenuation length and low light yield of these materials present challenges for their application in PET imaging.

An ideal, though costly, material is Lanthanum Bromide (LaBr3), which offers an exceptionally high light yield (63 photons/keV), fast timing (20 ns), and short attenuation length, all of which contribute to increased sensitivity. In contrast, Bismuth Germanate (BGO) has much longer decay times of 300 nanoseconds and a lower light yield of 8.5 photons per keV. Nevertheless, it is a more cost-effective alternative, allowing for the construction of longer scanners, thicker crystals, or a combination of both. Additionally, with its Cherenkov capabilities, BGO has the potential to deliver high-performance TOF PET detectors.

Depth of Interaction:

Regardless of the material used, increasing the crystal thickness must be carefully assessed, as it can lead to increased parallax effects in both radial and axial directions unless depth-of-interaction (DOI) information is also acquired[70]. DOI encodes the position inside the crystal where the incident 511-keV photon interacted with an electron in the scintillator material.

Quo Vadis, Total-Body PET?

Both the scanner length and TOF have a significant impact on the sensitivity in LAFOV and TB PET imaging. With TOF resolution continually improving and now reaching below 200 ps [12], [16], it is evident that TOF will be a key factor in future design considerations. Furthermore, the continuous TOF improvements over the last few years suggest that shorter PET scanners might achieve comparable effective sensitivity to current LAFOV PET scanners in the near future. For instance, while the uEXPLORER demonstrates a sensitivity approximately 8.7 times higher than the uMI Panorama, this comparison does not account for the effective sensitivity gain provided by superior TOF resolution. As a result, scanners without TB coverage could potentially suffice for most oncological and diagnostic tasks. For many imaging institutes the larger expenses for TB and LAFOV scanners compared to conventional PET scanners might make their procurement economically unfeasible. On the other hand, the potential to reduce the scan time could increase the patient throughput, while maintaining high diagnostic quality. The question of the optimal scanner length remains unresolved, given the vast parameter space defined by the number of readout channels, crystal type and thickness, and the availability of DOI information. All these factors are closely linked to costs associated with procurement, maintenance, and computing power. However, a trend is already emerging: following the uEXPLORER, the first true 2-meter-long total-body PET scanner, recent commercially available devices have reduced in length to 148 cm (uMI Panorama GS) and 106 cm (Biograph Vision Quadra). Ultimately, clinical applications will drive the decision-making process, necessitating a careful justification of total-body PET’s relevance based on clinical needs. A balanced trade-off between clinical benefits and the associated costs and space requirements must be carefully evaluated.

In the future, we may see the emergence of different “optimal” scanner lengths, crystal sizes and TOF resolutions tailored to specific applications. It is likely that no single scanner will be suitable for all purposes: oncology may find shorter scanners sufficient, while dynamic imaging studies require total-body coverage. Organ specific imagers like dedicated brain PET scanners might only need a short FOV but require smaller crystals and DOI information for massively improved spatial resolution[71]–better than needed for whole body scanners. Ergo, the driving force for the choice of crystal size and thickness will be the specific application.

6. Is affordable total-body PET imaging possible?

In parallel to the efforts of improving current scanner performance, the aspect of cost-effective and affordable imaging devices is increasingly investigated. Advancements in cost-effective PET scanner designs have the potential to dramatically improve access to high-quality medical imaging in low- and middle-income countries (LMICs), where current technology may be prohibitively expensive. By lowering the cost barrier, more institutions worldwide could adopt these technologies. This would effectively lead to a reduction of global health disparities, would initiate more international collaboration, could accelerate innovation, and would ensure that these advancements are not confined to wealthier nations. Several innovative approaches are currently under investigation to address this need.

At Jagiellonian University in Krakow, Poland, the development of the 200-cm long J-PET scanner represents an attempt to build a total-body PET scanner using inexpensive plastic scintillator strips. Simulation studies have demonstrated excellent timing performance and NECR, though with lower image contrast[13]. Due to its limited resolution – especially in the axial direction, where the resolution reaches about 7 mm – the J-PET TB scanner may not be suitable for standard oncologic diagnostic. However, it may still be viable for dynamic imaging and for obtaining image-derived input functions for kinetic modeling using the ascending aorta, which typically aligns with the scanner axis and is therefore less affected by low spatial resolution. A shorter prototype with a 50-cm axial FOV has already been built and is currently undergoing evaluation [72], [73], [74].

To improve patient throughput and address the spatial constraints of TB PET scanners, Vandenberghe et al. have proposed a cost-effective walk-through PET (WT-PET) scanner, which scans patients in an upright position [75]. The procurement costs for this scanner are estimated to be approximately 3.3 times lower than those for the Biograph Vision Quadra, with a projected increase in patient throughput by up to 1.6 times. Since CT imaging is required for scatter and attenuation correction, a standing CT could be integrated in line with the WT-PET system.

Additionally, sparse detector designs are being explored with the aim of maintaining a long axial FOV or even achieving total-body coverage while reducing the costs associated with scintillator crystals and readout channels [76], [77], [78]. The design configurations range from sparse detector block rings to a checkerboard arrangement and have so far only been investigated using Monte Carlo-methods. These designs, while more cost-effective, come with lower sensitivity, reduced spatial resolution, and still require substantial space. Nevertheless, sparse configurations may offer a clinically adoptable solution for cost-effective LAFOV or TB PET.

While cost-effective designs for TB and LAFOV PET scanners may be appealing for LMICs, it is important to consider the availability of radionuclides and radiochemical laboratories as an additional cost factor independent of the installed scanner. Moreover, the demand for LAFOV and TB PET systems is closely tied to the specific clinical tasks they are intended to address. The demand for affordable nuclear imaging devices may grow if many clinical applications requiring total-body coverage become routinely performed.

7. Large data handling and PHI

The progress of PET scanner technology brings about several benefits, including improvement in image quality and the capability to assess and qualify both pathologic and physiologic activities; on the other hand, several concerns have become apparent. These concerns include the need for safeguards and privacy protection protocols for the handling and using TB PET data. In addition to developing larger storage spaces and powerful machines to handle the data, we need to characterize and understand the many ways it might be utilized and the associated misuse. The data from conventional modalities has been protected to prevent patient identity and health information using anonymization of metainformation, which has worked for partial body imaging. Still, it may not be adequate for TB PET data. This paragraph will focus on the potential privacy and safety issues surrounding using TB PET data and the solutions being researched.

The data provided by tomographic scanners may be classified into physical, physiological, and behavioral categories, amongst others. It is easy to see that physical features are more explicit from TB PET images; for instance, weight, sex, body shape, hand shape, and, in some cases, facial features. The shape of the ears, nose, or the number of fingers may provide some clues when trying to identify a person [79]. Other less explicit physical biometrics can be more individual-specific, like a fingerprint, e.g., the vasculature patterns. It is easy to see that TB PET scans obtained with the current imaging standards may, in the future, be used to recognize individuals. With further advancement of the technology, the shape and movement of vocal cords might also be used to reproduce the voices of scanned individuals. Beyond anatomical information, a TB PET scan can provide functional physiological information. A dynamic scan can provide, for example, heart rate and potentially glomerular renal filtration rate, which can be combined with other data to tailor a personal medicine solution. Further analytical development of total body data can lead to recognizing behavioral patterns. These include less specific and readily available patterns/associations such as lifestyle choices.

It is now critical to carefully determine how data are stored and shared. While under current policies, a data set of PET/CT images without any of the information protected by the Health Insurance Portability and Accountability Act (HIPAA) is considered deidentified, this modern dataset could potentially identify a narrow group of individuals. Our group already brought up some of these concerns. For example, the study from Selfridge et al, initially addressed some of these concerns and provided a possible workaround [80]. However, they focused on readily available information (e.g. facial recognition) and did not focus on the future possibilities for a comprehensive total body PET/CT dataset. As an example, consider a volunteer that undergoes a total body 18F-FGD PET/CT. This person allows the research team to publicly share their deidentified (under HIPAA regulation) imaging dataset to an NIH database. In the coming years, anyone who accesses the NIH dataset containing images of an individual may not only be able to identify the person more easily using publicly available pictures on social media or other data unwittingly shared with other entities, but they could also collect crucial information about the person’s current and possibly future health conditions. This information could be used against the individual. At present, data such as this is used for research to understand the parameter space of TB PET, such as decreasing dose, acquisition times, or increasing uptake phase and image quality, which is the tip of the information iceberg. The future of TB PET research should include efforts to understand the extent and use of TB PET as it relates to individual privacy and national policy on data sharing should be informed by what is learned.

8. Conclusion

In summary, the evolution of LAFOV and TB PET scanners has significantly advanced the field of nuclear medicine, offering unprecedented improvements in sensitivity, TOF resolution, and image quality. However, the development of these systems also presents challenges, particularly regarding cost, and space requirements. As TOF resolution continues to improve, it is anticipated that shorter PET scanners might achieve comparable effective sensitivity to their longer counterparts, potentially reshaping the landscape of PET development. The future of PET technology will likely see a diverse array of scanner designs, each optimized for specific clinical applications, ranging from organ-specific imaging to dynamic studies requiring total-body coverage.

Looking ahead, we predict that the continuous enhancement of TOF resolution will push the boundaries of PET imaging, possibly paving the way for “reconstruction-free” imaging. Additionally, the drive towards cost-effective solutions will be crucial, particularly in making advanced PET imaging accessible worldwide. The integration of PET with other imaging modalities and the development of novel imaging agents will further expand the clinical utility of PET, supporting its growing role in personalized medicine. As these advancements unfold, careful consideration of regulatory, ethical, and logistical challenges will be necessary to ensure that the benefits of PET imaging are realized globally. Ultimately, the ongoing research and innovation in PET technology promise to enhance diagnostic capabilities, improve patient outcomes, and contribute to the broader goal of equitable healthcare access worldwide.

Abbreviations:

AFOV

axial field-of-view

FBP

forward-back-projection

FWHM

full width at half maximum

LORs

lines-of-response

LYSO

lutetium yttrium oxyorthosilicate

PSF

point-spread-function