Raman Spectroscopy and Machine Learning Enables Estimation of Articular Cartilage Structural, Compositional, and Functional Properties

OBJECTIVE

To differentiate healthy from artificially degraded articular cartilage and estimate its structural, compositional, and functional properties using Raman spectroscopy (RS).

DESIGN

Visually normal bovine patellae (n = 12) were used in this study. Osteochondral plugs (n = 60) were prepared and artificially degraded either enzymatically (via Collagenase D or Trypsin) or mechanically (via impact loading or surface abrasion) to induce mild to severe cartilage damage; additionally, control plugs were prepared (n = 12). Raman spectra were acquired from the samples before and after artificial degradation. Afterwards, reference biomechanical properties, proteoglycan (PG) content, collagen orientation, and zonal (%) thickness of the samples were measured. Machine learning models (classifiers and regressors) were then developed to discriminate healthy from degraded cartilage based on their Raman spectra and to predict the aforementioned reference properties.

RESULTS

The classifiers accurately categorized healthy and degraded samples (accuracy = 86%), and successfully discerned moderate from severely degraded samples (accuracy = 90%). On the other hand, the regression models estimated cartilage biomechanical properties with reasonable error (≤ 24%), with the lowest error observed in the prediction of instantaneous modulus (12%). With zonal properties, the lowest prediction errors were observed in the deep zone, i.e., PG content (14%), collagen orientation (29%), and zonal thickness (9%).

CONCLUSION

RS is capable of discriminating between healthy and damaged cartilage, and can estimate tissue properties with reasonable errors. These findings demonstrate the clinical potential of RS.

Assessment of ejection fraction and heart perfusion using myocardial perfusion single-photon emission computed tomography in Finland and Estonia: a multicenter phantom study

OBJECTIVES

Myocardial SPECT/CT imaging is frequently performed to assess myocardial perfusion and dynamic parameters of heart function, such as ejection fraction (EF). However, potential pitfalls exist in the imaging chain that can unfavorably affect diagnosis and treatment. We performed a national cardiac quality control study to investigate how much SPECT/CT protocols vary between different nuclear medicine units in Finland, and how this may affect the heart perfusion and EF values.

METHODS

Altogether, 21 nuclear medicine units participated with 27 traditional SPECT/CT systems and two cardiac-centered IQ-SPECT systems. The reproducibility of EF and the uniformity of perfusion were studied using a commercial dynamic heart phantom. SPECT/CT acquisitions were performed and processed at each participating unit using their own clinical protocol and with a standardized protocol. The effects of acquisition protocols and analysis routines on EF estimates and uniformity of perfusion were studied.

RESULTS

Considerable variation in EF estimates and in the uniformity of perfusion were observed between the units. Uniformity of perfusion was improved in some units after applying the higher count-statistic standard acquisition protocol. EF estimates varied more due to differences in analysis routines than as a result of different acquisition protocols. The results obtained with the two IQ-SPECT systems differed substantially from the traditional multipurpose cameras.

CONCLUSION

On average, the EF and heart perfusion were accurately estimated by SPECT/CT, but high errors could be produced if the acquisition and analysis routines were poorly optimized. Eight of the 21 participants altered their imaging protocol after this quality control tour.