Parallel Data Aggregation for PET Event Loss Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Positron emission tomography (PET) scanners experience significant event losses due to the daisy-chained data transmission scheme, where only one event can be transmitted per sync period, leading to discarded events, especially at higher count rates.
Innovation Solution
Implementing a parallel data aggregation scheme with field-programmable gate arrays (FPGAs) to connect each detector data slot directly to a control interface, allowing multiple events to be transmitted simultaneously during a sync period, thereby reducing event losses and optimizing data bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a daisy-chained data transmission scheme is used to connect detector slots in series, then device complexity is reduced and ease of manufacture is improved, but event losses increase significantly at higher count rates because only one event can be transmitted per sync period
Solution Approach 1:
The patent divides the data transmission path into parallel segments, where each detector slot connects to its own dedicated buffer memory rather than being daisy-chained. This segmentation allows multiple events from different slots to be captured simultaneously in separate buffers, eliminating the bottleneck of series connection while maintaining manufacturing simplicity through modular buffer units.
Solution Approach 2:
The patent transitions from a one-dimensional series connection to a two-dimensional parallel architecture by introducing a time dimension through buffer memories. Events are now stored in buffers across multiple time cycles before being transmitted, adding a temporal dimension to the data flow that allows simultaneous capture of multiple events without increasing physical connection complexity.
2Device complexity
If a daisy-chained data transmission scheme is used, then device complexity is reduced, but productivity decreases because events are discarded when multiple events occur in the same sync period
Solution Approach 1:
The patent implements preliminary action by providing buffer memories at each detector slot that can pre-capture events before the transmission bottleneck occurs. These buffers hold events temporarily, allowing the system to accumulate data from multiple slots over several sync periods and then transmit them in batches, thereby increasing effective data capture without requiring complex real-time multiplexing hardware.
Solution Approach 2:
The buffer memory acts as an intermediary between the detector slots and the transmission bus. It decouples the data capture function from the data transmission function, allowing events to be recorded independently of transmission timing constraints. This intermediary buffer resolves the conflict between simple series transmission and high productivity by absorbing the temporal mismatch between event occurrence and data transmission.
3Loss of information
If multiple events are captured simultaneously in parallel buffers, then event losses are reduced and noise equivalent count rate increases by 10-15%, but device complexity increases due to additional buffer memories and control logic
Solution Approach 1:
The patent merges the buffer memory functions into a unified transmission mechanism where all buffered events from parallel slots are combined and transmitted together through the existing data bus infrastructure. This merging approach allows the system to achieve parallel data capture capabilities while utilizing shared transmission resources, thereby reducing event losses without proportionally increasing overall system complexity.
Solution Approach 2:
The patent changes the temporal parameter of data transmission by introducing buffer depth as a controllable parameter. By adjusting buffer size and transmission timing, the system can optimize the balance between capturing multiple simultaneous events and managing hardware complexity. The buffer acts as a parameter-controlled intermediary that can be tuned to match specific operational requirements without fundamentally altering the core architecture.
Data Source
AI summary
An event data transmission scheme is provided for reducing positron emission tomography event losses. The event data transmission scheme employs a more effective use of available data bandwidth. Each of a plurality of detector data slots is connected directly to a data aggregation control interface, and the control interface is connected to a coincidence processor.