Keith Derek Rogers

Background/Objectives: Establishing biomarkers is crucial for early cancer detection. Structural biomarkers can be revealed by noninvasive X-ray diffraction. We applied this approach to human nails and observed structural changes in the major nail tissue component, keratin, which coincided with a cancer diagnosis. Methods: X-ray diffraction patterns were obtained from 230 nail samples (control/cancer split: 92/138) belonging to 133 patients (48/85) in wide (WAXS) and small (SAXS) ranges of scattering angles. For both ranges, we obtained one-dimensional scattering profiles in meridional and equatorial directions. We examined these profiles using machine-learning techniques such as logistic regression and linear discriminant analysis. Results: We obtained good classification metrics that allow us to distinguish participants with cancer from controls. The results are similar for both machine-learning methods. Combining all four profiles (WAXS meridional, WAXS equatorial, SAXS meridional, and SAXS equatorial) produced a sensitivity of 0.753 and a specificity of 0.771 for the Linear Discriminant Analysis at the Youden point. For the limited set of patients of the same age, we showed that the patterns can still be distinguished, supporting our hypothesis that the observed aberrations are cancer-induced. Conclusions: Our exploratory research demonstrated the feasibility of cancer detection using the X-ray diffraction of human nails.

Background/Objectives: Establishing biomarkers is crucial for early cancer detection. Structural biomarkers can be revealed by noninvasive X-ray diffraction. We applied this approach to human nails and observed cancer-induced structural changes in the major nail tissue component, keratin. Methods: X-ray diffraction patterns were obtained from 230 nail samples (control/cancer split: 92/138) belonging to 133 patients (48/85) in wide (WAXS) and small (SAXS) ranges of scattering angles. For both ranges, we obtained one-dimensional scattering profiles in meridional and equatorial directions. We examined these profiles using machine-learning techniques such as logistic regression and linear discriminant analysis. Results: We obtained significant classification metrics that allow us to distinguish healthy and cancerous patients. The results are similar for both machine-learning methods. Combining all four profiles (WAXS meridional, WAXS equatorial, SAXS meridional, and SAXS equatorial) produced a sensitivity of 0.753 and a specificity of 0.771. For the limited set of patients of the same age, we showed that the patterns can still be distinguished, supporting our hypothesis that the observed aberrations are cancer-induced. Conclusions: We showed that cancer detection can be achieved by means of X-ray diffraction of human nails.

Background/Objectives: Machine learning approaches are widely used in modern medical diagnostics, including cancer detection. The results can be significantly improved by aggregating individual measurements, and appropriate aggregation methods should be established. Methods: We applied various measurement aggregation strategies both before and after machine learning modeling to two datasets of X-ray diffraction images: human breast biopsy samples and canine claw samples. Two classifiers, Random Forest and Logistic Regression, were used to determine classification metrics: the area under the receiver operating characteristic curve (ROC-AUC) and balanced accuracy. Results: We found that all aggregation types improve classification metrics, with aggregation after modeling yielding better performance. Depending on the dataset and approach, either classifier can produce better results. For human breast samples, Random Forest with the logit aggregation strategy provides an ROC-AUC exceeding 0.9. For the canine dataset, both Random Forest with the logit aggregation strategy and Logistic Regression with the median of cancer probabilities achieve an ROC-AUC of about 0.85. Conclusions: We examined several simple, straightforward aggregation methods for patient diagnosis based on multiple measurements per patient and achieved significant improvements in classification metrics.

Structural biomarkers determined by X-ray scattering of the tissues can complement conventional histopathology and facilitate a fast triage procedure of cancer biopsy samples. It has been shown previously that lipid reflexes can distinguish cancerous from benign samples, except for fibroadenomas. In the present study, we demonstrate that fibroadenoma samples can be recognized using X-ray scattering of collagen. Moreover, we show that modifications in collagen structure are manifested in the water reflexes. Examination of diffraction patterns from water using two-dimensional Fourier transformation and machine learning yields excellent classification metrics in both synchrotron images and laboratory diffractometer data.

Structural biomarkers determined by X-ray scattering of the tissues can complement conventional histopathology and facilitate a fast triage procedure of cancer biopsy samples. It has been shown previously that lipid reflexes can distinguish cancerous from benign samples, except for fibroadenomas. In the present study, we demonstrate that fibroadenoma samples can be recognized using X-ray scattering of collagen. Moreover, we show that modifications in collagen structure are manifested in the water reflexes. Examination of diffraction patterns from water using two-dimensional Fourier transformation and machine learning yields excellent classification metrics in both synchrotron images and laboratory diffractometer data.

Mammographic radiomic features were able to predict occult invasive disease in core-needle biopsy–proven ductal carcinoma in situ in patients from three countries.

New methods for real-time materials phase identification based upon focal construct tomography (FCT) have been examined. Such quantitative assessment has significant potential in sectors where in-line analysis is required, including screening within aviation security. As a recent component of work programs developing FCT, its capability for accurate, quantitative analysis has been assessed for the first time. Diffraction signatures from mixed-phase materials were acquired from an energy-dispersive FCT system running under normal operational conditions. A calibration curve was constructed from the spectra and subsequently employed to assess the composition of ‘blind’ samples. The results demonstrated that this approach was able to precisely predict the polymorphic phase composition of samples to ±5 wt%. Conclusions: The potential impact of these findings is significant and will enable applications of FCT beyond those requiring a phase identification to those necessitating quantification, such as counterfeit medicines, pharmaceutical quality assurance, aging of explosives, and cement production.

Structural biomarkers determined by X-ray scattering of the tissues can complement conventional diagnostics and provide a pathway for early detection of diseases. In the present study, mouse models were utilized to observe the progression of prostate cancer. We induced cancer in the left lobe of the mouse prostate, whilst the right lobe was left uninoculated. The mice were sacrificed at increasing systematic time points, and lobe samples were subsequently analyzed using X-ray scattering. Control samples were also collected from healthy mice sacrificed at the same time points. This investigation revealed that the ratio between the X-ray scattering peaks associated with the lipids and water can serve as a structural biomarker of cancer, and this biomarker develops as the tumor advances. The obtained cancer trajectory can serve as a baseline for the determination of the disease stage, and the biomarker movement along the trajectory can be evidence of the healing or disease progression.

Background/Objectives: With the number of detected breast cancer cases growing every year, there is a need to augment histopathological analysis with fast preliminary screening. We examine the feasibility of using X-ray diffraction measurements for this purpose. Methods: In this work, we obtained more than 6000 diffraction patterns from 211 patients and examined both standard and custom-developed methods, including Fourier coefficient analysis, for their interpretation. Various preprocessing steps and machine learning classifiers were compared to determine the optimal combination. Results: We demonstrated that benign and cancerous clusters are well separated, with specificity and sensitivity exceeding 0.9. For wide-angle scattering, the two-dimensional Fourier method is superior, while for small angles, the conventional analysis based on azimuthal integration of the images provides similar metrics. Conclusions: X-ray diffraction of biopsy tissues, supported by machine learning approaches to data analytics, can be an essential tool for pathological services. The method is rapid and inexpensive, providing excellent metrics for benign/cancer classification.

Background/Objectives: With the number of detected breast cancer cases growing every year, there is a need to augment histopathological analysis with fast preliminary screening. We examine the feasibility of using X-ray diffraction measurements for this purpose; Methods: In this work, we obtained more than 6,000 diffraction patterns from 211 patients and examined both standard and custom-developed methods, including Fourier coefficient analysis, for their interpretation. Various preprocessing steps and machine-learning classifiers were compared to determine the optimal combination; Results: We demonstrated that benign and cancerous clusters are well-separated, with specificity and sensitivity exceeding 0.9. For wide-angle scattering, the two-dimensional Fourier method is superior, while for small angles, the conventional analysis based on azimuthal integration of the images provides similar metrics; Conclusions: X-ray diffraction of biopsy tissues, supported by machine-learning approaches to data analytics, can be an essential tool for pathological services. The method is rapid and inexpensive, providing excellent metrics for benign/cancer classification.