Early COVID-19 respiratory risk stratification using machine learning.

Background: COVID-19 has strained healthcare systems globally. In this and future pandemics, providers with limited critical care experience must distinguish between moderately ill patients and those who will require aggressive care, particularly endotracheal intubation. We sought to develop a machine learning-informed Early COVID-19 Respiratory Risk Stratification (ECoRRS) score to assist in triage, by providing a prediction of intubation within the next 48 hours based on objective clinical parameters. Methods: Electronic health record data from 3447 COVID-19 hospitalizations, 20.7% including intubation, were extracted. 80% of these records were used as the derivation cohort. The validation cohort consisted of 20% of the total 3447 records. Multiple randomizations of the training and testing split were used to calculate confidence intervals. Data were binned into 4-hour blocks and labeled as cases of intubation or no intubation within the specified time frame. A LASSO (least absolute shrinkage and selection operator) regression model was tuned for sensitivity and sparsity. Results: Six highly predictive parameters were identified, the most significant being fraction of inspired oxygen. The model achieved an area under the receiver operating characteristic curve of 0.789 (95% CI 0.785 to 0.812). At 90% sensitivity, the negative predictive value was 0.997. Discussion: The ECoRRS score enables non-specialists to identify patients with COVID-19 at risk of intubation within 48 hours with minimal undertriage and enables health systems to forecast new COVID-19 ventilator needs up to 48 hours in advance. Level of evidence: IV.

Altered nuclear architecture in blood cells from Huntington's disease patients.

BACKGROUND: Cell nuclear architecture has been explored in cancer and laminopathies but not in neurodegenerative disorders. Huntington's disease (HD) is a neurodegenerative disorder that leads to neuronal death. Chromosome-wide changes in gene expression have been reported in HD, not only in the brain but also in peripheral blood cells, but whether this translates into nuclear and chromosome architecture alterations has not yet been studied. METHODS: We investigate nuclear structure and chromosome organization in HD blood cells using fluorescence in situ hybridization in ultrathin cryosections (cryoFISH), coupled with machine learning image analysis to evaluate size, distribution, and morphology of nuclei and chromosomes. Four chromosomes were analyzed based on up- or downregulation of gene expression in HD. RESULTS: We show that blood cells from HD patients display increased nuclear size and filamentary shape, increased size of gene-rich chromosome 19, decreased filamentary shape of gene-rich chromosome 22, and a more radially centralized position for chromosome 19, whereas chromosomes 4 and 5 do not show detectable differences. CONCLUSIONS: We identify gross changes in nuclear architecture and chromosome organization associated with HD in blood. This adds a new layer of information onto disrupting mechanisms in HD and increases the potential of using blood to survey HD.

Loss of Mammographic Tissue Homeostasis in Invasive Lobular and Ductal Breast Carcinomas vs. Benign Lesions.

The 2D wavelet transform modulus maxima (WTMM) method is used to perform a comparison of the spatial fluctuations of mammographic breast tissue from patients with invasive lobular carcinoma, those with invasive ductal carcinoma, and those with benign lesions. We follow a procedure developed and validated in a previous study, in which a sliding window protocol is used to analyze thousands of small subregions in a given mammogram. These subregions are categorized according to their Hurst exponent values (H): fatty tissue (H ≤ 0.45), dense tissue (H ≥ 0.55), and disrupted tissue potentially linked with tumor-associated loss of homeostasis (0.45 < H < 0.55). Following this categorization scheme, we compare the mammographic tissue composition of the breasts. First, we show that cancerous breasts are significantly different than breasts with a benign lesion (p-value ∼ 0.002). Second, the asymmetry between a patient's cancerous breast and its contralateral counterpart, when compared to the asymmetry from patients with benign lesions, is also statistically significant (p-value ∼ 0.006). And finally, we show that lobular and ductal cancerous breasts show similar levels of disruption and similar levels of asymmetry. This study demonstrates reproducibility of the WTMM sliding-window approach to help detect and characterize tumor-associated breast tissue disruption from standard mammography. It also shows promise to help with the detection lobular lesions that typically go undetected via standard screening mammography at a much higher rate than ductal lesions. Here both types are assessed similarly.

Using convolutional neural networks to discriminate between cysts and masses in Monte Carlo-simulated dual-energy mammography.

PURPOSE: A substantial percentage of recalls (up to 20%) in screening mammography is attributed to extended round lesions. Benign fluid-filled breast cysts often appear similar to solid tumors in conventional mammograms. Spectral imaging (dual-energy or photon-counting mammography) has been shown to discriminate between cysts and solid masses with clinically acceptable accuracy. This work explores the feasibility of using convolutional neural networks (CNNs) for this task. METHODS: A series of Monte Carlo experiments was conducted with digital breast phantoms and embedded synthetic lesions to produce realistic dual-energy images of both lesion types. We considered such factors as nonuniform anthropomorphic background, size of the mass, breast compression thickness, and variability in lesion x-ray attenuation. These data then were used to train a deep neural network (ResNet-18) to learn the differences in x-ray attenuation of cysts and masses. RESULTS: Our simulation results showed that the CNN-based classifier could reliably discriminate between cystic and solid mass round lesions in dual-energy images with an area under the receiver operating characteristic curve (ROC AUC) of 0.98 or greater. CONCLUSIONS: The proposed approach showed promising performance and ease of implementation, and could be applied to novel photon-counting detector-based spectral mammography systems.

Mammographic evidence of microenvironment changes in tumorous breasts.

PURPOSE: The microenvironment of breast tumors plays a critical role in tumorigenesis. As long as the structural integrity of the microenvironment is upheld, the tumor is suppressed. If tissue structure is lost through disruptions in the normal cell cycle, the microenvironment may act as a tumor promoter. Therefore, the properties that distinguish between healthy and tumorous tissues may not be solely in the tumor characteristics but rather in surrounding non-tumor tissue. The goal of this paper was to show preliminary evidence that tissue disruption and loss of homeostasis in breast tissue microenvironment and breast bilateral asymmetry can be quantitatively and objectively assessed from mammography via a localized, wavelet-based analysis of the whole breast. METHODS: A wavelet-based multifractal formalism called the 2D Wavelet Transform Modulus Maxima (WTMM) method was used to quantitate density fluctuations from mammographic breast tissue via the Hurst exponent (H). Each entire mammogram was cut in hundreds of 360 × 360 pixel subregions in a gridding scheme of overlapping sliding windows, with each window boundary separated by 32 pixels. The 2D WTMM method was applied to each subregion individually. A data mining approach was set up to determine which metrics best discriminated between normal vs. cancer cases. These same metrics were then used, without modification, to discriminate between normal vs. benign and benign vs. cancer cases. RESULTS: The density fluctuations in healthy mammographic breast tissue are either monofractal anti-correlated (H < 1/2) for fatty tissue or monofractal long-range correlated (H>1/2) for dense tissue. However, tissue regions with H~1/2, as well as left vs. right breast asymetries, were found preferably in tumorous (benign or cancer) breasts vs. normal breasts, as quantified via a combination metric yielding a P-value ~ 0.0006. No metric considered showed significant differences between cancer vs. benign breasts. CONCLUSIONS: Since mammographic tissue regions associated with uncorrelated (H~1/2) density fluctuations were predominantly in tumorous breasts, and since the underlying physical processes associated with a H~1/2 signature are those of randomness, lack of spatial correlation, and free diffusion, it is hypothesized that this signature is also associated with tissue disruption and loss of tissue homeostasis.

Comparative Multifractal Analysis of Dynamic Infrared Thermograms and X-Ray Mammograms Enlightens Changes in the Environment of Malignant Tumors.

There is growing evidence that the microenvironment surrounding a tumor plays a special role in cancer development and cancer therapeutic resistance. Tumors arise from the dysregulation and alteration of both the malignant cells and their environment. By providing tumor-repressing signals, the microenvironment can impose and sustain normal tissue architecture. Once tissue homeostasis is lost, the altered microenvironment can create a niche favoring the tumorigenic transformation process. A major challenge in early breast cancer diagnosis is thus to show that these physiological and architectural alterations can be detected with currently used screening techniques. In a recent study, we used a 1D wavelet-based multi-scale method to analyze breast skin temperature temporal fluctuations collected with an IR thermography camera in patients with breast cancer. This study reveals that the multifractal complexity of temperature fluctuations superimposed on cardiogenic and vasomotor perfusion oscillations observed in healthy breasts is lost in malignant tumor foci in cancerous breasts. Here we use a 2D wavelet-based multifractal method to analyze the spatial fluctuations of breast density in the X-ray mammograms of the same panel of patients. As compared to the long-range correlations and anti-correlations in roughness fluctuations, respectively observed in dense and fatty breast areas, some significant change in the nature of breast density fluctuations with some clear loss of correlations is detected in the neighborhood of malignant tumors. This attests to some architectural disorganization that may deeply affect heat transfer and related thermomechanics in breast tissues, corroborating the change to homogeneous monofractal temperature fluctuations recorded in cancerous breasts with the IR camera. These results open new perspectives in computer-aided methods to assist in early breast cancer diagnosis.

Computational growth model of breast microcalcification clusters in simulated mammographic environments.

BACKGROUND: When screening for breast cancer, the radiological interpretation of mammograms is a difficult task, particularly when classifying precancerous growth such as microcalcifications (MCs). Biophysical modeling of benign vs. malignant growth of MCs in simulated mammographic backgrounds may improve characterization of these structures METHODS: A mathematical model based on crystal growth rules for calcium oxide (benign) and hydroxyapatite (malignant) was used in conjunction with simulated mammographic backgrounds, which were generated by fractional Brownian motion of varying roughness and quantified by the Hurst exponent to mimic tissue of varying density. Simulated MC clusters were compared by fractal dimension, average circularity of individual MCs, average number of MCs per cluster, and average cluster area. RESULTS: Benign and malignant clusters were distinguishable by average circularity, average number of MCs per cluster, and average cluster area with p<0.01 across all Hurst exponent values considered. Clusters were distinguishable by fractal dimension with p<0.05 in low Hurst exponent environments. As the Hurst exponent increased (tissue density increased) benign and malignant MCs became indistinguishable by fractal dimension. CONCLUSIONS: The fractal dimension of MCs changes with breast tissue density, which suggests tissue environment plays a role in regulating MC growth. Benign and malignant MCs are distinguishable in all types of tissue by shape, size, and area, which is consistent with findings in the literature. These results may help to better understand the effects of the tissue environment on tumor progression, and improve classification of MCs in mammograms via computer-aided diagnosis.