Updated: Nov 9, 2025

Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
Published on: June 10, 2025
D Cody Morris1, Derek Y Chan1, Mark L Palmeri1
1Department of Biomedical Engineering, Duke University, Durham, North Carolina, USA.
You might also read
Articles linked to this work by shared authors, journal, and citation graph.
This study introduces a new 3D ultrasound technique to better identify prostate cancer. By measuring tissue stiffness, the method helps doctors locate tumors more accurately than standard ultrasound, potentially improving biopsy results.
Area of Science:
Background:
Standard ultrasound often lacks the precision required for reliable prostate cancer identification during biopsy procedures. This limitation creates a significant obstacle for clinicians seeking to target malignant lesions effectively. Prior research has shown that conventional imaging methods frequently miss small or subtle tumors within the gland. That uncertainty drove the development of advanced elasticity-based diagnostic tools. No prior work had resolved how to capture volumetric stiffness data across the entire organ efficiently. This gap motivated the creation of a system utilizing acoustic radiation force to generate shear waves. Researchers aimed to improve signal quality compared to existing clinical platforms. These efforts provide a foundation for evaluating tissue mechanical properties in a three-dimensional space.
Purpose Of The Study:
The aim of this study was to evaluate the feasibility of using three-dimensional shear wave elasticity imaging for detecting and localizing prostate cancer. Researchers sought to overcome the insufficient sensitivity associated with traditional transrectal ultrasound. They hypothesized that a dense acoustic radiation force push would enhance signal-to-noise ratios. This motivation drove the development of a system capable of screening the entire gland. The team intended to establish specific stiffness thresholds to differentiate malignant tissue from healthy structures. They also aimed to determine if normalizing for tissue compression could reduce patient variability. This objective addressed the need for more reliable guidance during biopsy procedures. The investigation ultimately focused on validating these metrics against clinical outcomes in patients undergoing surgery.
The researchers propose that 3D shear wave speed measurements identify cancer by detecting increased tissue stiffness. They established a threshold of 5.6 m/s, or 94.1 kPa, to differentiate malignant regions from healthy prostate tissue with an area under the curve of 0.84.
The system utilizes a dense acoustic radiation force push approach. This specific configuration generates shear waves that provide a higher signal-to-noise ratio than current commercial platforms, allowing for the comprehensive screening of the entire gland before performing a biopsy.
A dense acoustic radiation force push is necessary to acquire volumetric data. This method facilitates the screening of the entire prostate gland, which is required to overcome the limitations of standard B-mode ultrasound that often fails to target lesions accurately.
Main Methods:
Review approach involved assessing thirty-six patients scheduled for radical prostatectomy using the novel volumetric system. Investigators captured three-dimensional data to evaluate mechanical properties across the entire gland. The team employed a dense acoustic radiation force push to generate signals. They calculated mean speeds in various anatomical zones to establish baseline tissue characteristics. Receiver operating characteristic analyses determined optimal thresholds for separating malignant from benign areas. The researchers also assessed a speed ratio to account for external compression effects. This normalization process aimed to reduce variability between different subjects during the examination. Finally, the group compared the diagnostic performance of these metrics against established clinical standards.
Main Results:
Key findings from the literature reveal that cancer exhibits a mean shear wave speed of 6.0 m/s, or 108.0 kPa. In contrast, the peripheral zone showed a mean speed of 4.8 m/s, while the central gland measured 5.3 m/s. Differences between these regions reached statistical significance with p-values below 0.0001. A speed threshold of 5.6 m/s identified cancer with 81% sensitivity and 82% specificity. Normalizing for compression improved the area under the curve to 0.90. This adjusted metric achieved 75% sensitivity and 90% specificity for tumor detection. The positive predictive value rose to 79% when using the normalized speed ratio. These results confirm the feasibility of using volumetric data for accurate tumor localization.
Conclusions:
The authors demonstrate that volumetric elasticity mapping successfully distinguishes malignant tissue from healthy prostate regions. Their findings suggest that mechanical stiffness thresholds provide diagnostic accuracy comparable to advanced magnetic resonance fusion techniques. Synthesis and implications indicate that normalizing for tissue compression significantly enhances the predictive power of these measurements. The researchers propose that this approach offers a viable pathway for improving biopsy targeting precision. Statistical analysis confirms that shear wave speed differences between healthy and cancerous zones remain highly significant. This work highlights the potential for integrating volumetric stiffness data into routine clinical workflows. The study confirms that specific speed ratios effectively account for patient-to-patient variability during examination. These results support the broader adoption of advanced ultrasound techniques for localized cancer detection.
The study uses 3D shear wave speed data to quantify tissue elasticity. By calculating a speed ratio, the researchers normalize for external compression and patient-specific variability, which improves the specificity and positive predictive value of the diagnostic results to 90% and 79% respectively.
The researchers measured mean shear wave speeds of 4.8 m/s in the peripheral zone, 5.3 m/s in the central gland, and 6.0 m/s for cancer. These values represent statistically significant differences, with p-values lower than 0.0001 across all assessed prostate regions.
The authors suggest that their 3D imaging approach is feasible for detecting and localizing tumors. They claim that normalizing for applied compression during data acquisition provides clear benefits for future biopsy targeting studies compared to non-normalized measurements.