Pulmonary emphysema subtypes defined by unsupervised machine learning on CT scans.

BACKGROUND: Treatment and preventative advances for chronic obstructive pulmonary disease (COPD) have been slow due, in part, to limited subphenotypes. We tested if unsupervised machine learning on CT images would discover CT emphysema subtypes with distinct characteristics, prognoses and genetic associations. METHODS: New CT emphysema subtypes were identified by unsupervised machine learning on only the texture and location of emphysematous regions on CT scans from 2853 participants in the Subpopulations and Intermediate Outcome Measures in COPD Study (SPIROMICS), a COPD case-control study, followed by data reduction. Subtypes were compared with symptoms and physiology among 2949 participants in the population-based Multi-Ethnic Study of Atherosclerosis (MESA) Lung Study and with prognosis among 6658 MESA participants. Associations with genome-wide single-nucleotide-polymorphisms were examined. RESULTS: The algorithm discovered six reproducible (interlearner intraclass correlation coefficient, 0.91-1.00) CT emphysema subtypes. The most common subtype in SPIROMICS, the combined bronchitis-apical subtype, was associated with chronic bronchitis, accelerated lung function decline, hospitalisations, deaths, incident airflow limitation and a gene variant near DRD1, which is implicated in mucin hypersecretion (p=1.1 ×10-8). The second, the diffuse subtype was associated with lower weight, respiratory hospitalisations and deaths, and incident airflow limitation. The third was associated with age only. The fourth and fifth visually resembled combined pulmonary fibrosis emphysema and had distinct symptoms, physiology, prognosis and genetic associations. The sixth visually resembled vanishing lung syndrome. CONCLUSION: Large-scale unsupervised machine learning on CT scans defined six reproducible, familiar CT emphysema subtypes that suggest paths to specific diagnosis and personalised therapies in COPD and pre-COPD.

Radiation maps of ocean sediment from the Castle Bravo crater.

On March 1, 1954, the United States conducted its largest thermonuclear weapon test in Bikini Atoll in the Marshall Islands; the detonation was code-named "Castle Bravo." Radioactive deposits in the ocean sediment at the bomb crater are widespread and high levels of contamination remain today. One hundred thirty cores were collected from the top 25 cm of surface sediment at ocean depths approaching 60 m over a ∼2-km2 area, allowing for a presentation of radiation maps of the Bravo crater site. Radiochemical analyses were performed on the following radionuclides: plutonium-(239,240), plutonium-238, americium-241, bismuth-207, and cesium-137. Large values of plutonium-(239,240), americium-241, and bismuth-207 are found. Comparisons are made to core sample results from other areas in the northern Marshall Islands.

Background gamma radiation and soil activity measurements in the northern Marshall Islands.

We report on measurements of external gamma radiation on 9 islands in 4 atolls in the northern Marshall Islands, all of which were affected by the US nuclear testing program from 1946 to 1958 (Enjebi, Ikuren, and Japtan in Enewetak Atoll; Bikini and Enyu in Bikini Atoll; Naen in Rongelap Atoll; and Aon, Elluk, and Utirik in Utirik Atoll). We also report americium-241, cesium-137, plutonium-238, and plutonium-239,240 activity concentrations in the soil samples for 11 islands in 4 northern atolls (Enewetak, Japtan, Medren, and Runit in Enewetak Atoll; Bikini and Enyu in Bikini Atoll; Naen and Rongelap in Rongelap Atoll; and Aon, Elluk, and Utirik in Utirik Atoll) and from Majuro Island, Majuro Atoll in the southern Marshall Islands. Our results show low external gamma radiation levels on some islands in the Enewetak Atoll and Utirik Atoll, and elevated levels on Enjebi Island in the Enewetak Atoll, on Bikini Atoll, and on Naen Island in the Rongelap Atoll. We perform ordinary kriging on external gamma radiation measurements to provide interpolated maps. We find that radionuclides are absent from all Majuro soil samples, and that they are present at highest activity concentrations in samples from Runit and Enjebi islands (Enewetak Atoll), Bikini Island (Bikini Atoll), and Naen Island (Rongelap Atoll). We contextualize all results by making comparisons between islands and to various standards, as well as to regions of the world affected by nuclear accidents. We also discuss implications for informed decision-making by the Marshallese and local atoll governments and their people on issues pertaining to island resettlement.

In situ measurement of cesium-137 contamination in fruits from the northern Marshall Islands.

Radioactive contamination of fruits in the northern Marshall Islands, resulting from the US nuclear weapons testing program in the 1940s and 1950s, is still a human health concern, in particular pertaining to island population resettlement and the economic benefit from farming. Over 200 fruits, primarily coconuts and pandanus, were collected on 11 islands from four atolls in the northern Marshall Islands in 2017. The energy spectra from nuclear gamma decays were measured on a research vessel for each fruit in situ. From these recordings, the level of cesium-137 (137Cs) contamination was determined for individual fruits. Comparisons of the results are made to past studies and international food safety standards. There is a broad distribution of values, ranging from below detectable radiation levels to relatively high levels; safety concerns are largest for Bikini Island. A noticeable fraction of fruits from Bikini have significantly higher levels of 137Cs contamination compared with those from all other measured islands.

Human airway branch variation and chronic obstructive pulmonary disease.

Susceptibility to chronic obstructive pulmonary disease (COPD) beyond cigarette smoking is incompletely understood, although several genetic variants associated with COPD are known to regulate airway branch development. We demonstrate that in vivo central airway branch variants are present in 26.5% of the general population, are unchanged over 10 y, and exhibit strong familial aggregation. The most common airway branch variant is associated with COPD in two cohorts (n = 5,054), with greater central airway bifurcation density, and with emphysema throughout the lung. The second most common airway branch variant is associated with COPD among smokers, with narrower airway lumens in all lobes, and with genetic polymorphisms within the FGF10 gene. We conclude that central airway branch variation, readily detected by computed tomography, is a biomarker of widely altered lung structure with a genetic basis and represents a COPD susceptibility factor.

Measurement of background gamma radiation in the northern Marshall Islands.

We report measurements of background gamma radiation levels on six islands in the northern Marshall Islands (Enewetak, Medren, and Runit onEnewetak Atoll; Bikini and Nam on Bikini Atoll; and Rongelap on Rongelap Atoll). Measurable excess radiation could be expected from the decay of (137)Cs produced by the US nuclear testing program there from 1946 to 1958. These recordings are of relevance to safety of human habitation and resettlement. We find low levels of gamma radiation for the settled island of Enewetak [mean = 7.6 millirem/year (mrem/y) = 0.076 millisievert/year (mSv/y)], larger levels of gamma radiation for the island of Rongelap (mean = 19.8 mrem/y = 0.198 mSv/y), and relatively high gamma radiation on the island of Bikini (mean = 184 mrem/y = 1.84 mSv/y). Distributions of gamma radiation levels are provided, and hot spots are discussed. We provide interpolated maps for four islands (Enewetak, Medren, Bikini, and Rongelap), and make comparisons to control measurements performed on the island of Majuro in the southern Marshall Islands, measurements made in Central Park in New York City, and the standard agreed upon by the United States and the Republic of the Marshall Islands (RMI) governments (100 mrem/y = 1 mSv/y). External gamma radiation levels on Bikini Island significantly exceed this standard (P = <<0.01), and external gamma radiation levels on the other islands are below the standard. To determine conclusively whether these islands are safe for habitation, radiation exposure through additional pathways such as food ingestion must be considered.

Mapping Alveolar Oxygen Partial Pressure in COPD Using Hyperpolarized Helium-3: The Multi-Ethnic Study of Atherosclerosis (MESA) COPD Study.

Chronic obstructive pulmonary disease (COPD) and emphysema are characterized by functional and structural damage which increases the spaces for gaseous diffusion and impairs oxygen exchange. Here we explore the potential for hyperpolarized (HP) 3He MRI to characterize lung structure and function in a large-scale population-based study. Participants (n = 54) from the Multi-Ethnic Study of Atherosclerosis (MESA) COPD Study, a nested case-control study of COPD among participants with 10+ packyears underwent HP 3He MRI measuring pAO2, apparent diffusion coefficient (ADC), and ventilation. HP MRI measures were compared to full-lung CT and pulmonary function testing. High ADC values (>0.4 cm2/s) correlated with emphysema and heterogeneity in pAO2 measurements. Strong correlations were found between the heterogeneity of global pAO2 as summarized by its standard deviation (SD) (p < 0.0002) and non-physiologic pAO2 values (p < 0.0001) with percent emphysema on CT. A regional study revealed a strong association between pAO2 SD and visual emphysema severity (p < 0.003) and an association with the paraseptal emphysema subtype (p < 0.04) after adjustment for demographics and smoking status. HP noble gas pAO2 heterogeneity and the fraction of non-physiological pAO2 results increase in mild to moderate COPD. Measurements of pAO2 are sensitive to regional emphysematous damage detected by CT and may be used to probe pulmonary emphysema subtypes. HP noble gas lung MRI provides non-invasive information about COPD severity and lung function without ionizing radiation.

Quantification of lung ventilation defects on hyperpolarized MRI: The Multi-Ethnic Study of Atherosclerosis (MESA) COPD study.

PURPOSE: To develop an end-to-end deep learning (DL) framework to segment ventilation defects on pulmonary hyperpolarized MRI. MATERIALS AND METHODS: The Multi-Ethnic Study of Atherosclerosis Chronic Obstructive Pulmonary Disease (COPD) study is a nested longitudinal case-control study in older smokers. Between February 2016 and July 2017, 56 participants (age, mean ± SD, 74 ± 8 years; 34 men) underwent same breath-hold proton (1H) and helium (3He) MRI, which were annotated for non-ventilated, hypo-ventilated, and normal-ventilated lungs. In this retrospective DL study, 820 1H and 3He slices from 42/56 (75%) participants were randomly selected for training, with the remaining 14/56 (25%) for test. Full lung masks were segmented using a traditional U-Net on 1H MRI and were imported into a cascaded U-Net, which were used to segment ventilation defects on 3He MRI. Models were trained with conventional data augmentation (DA) and generative adversarial networks (GAN)-DA. RESULTS: Conventional-DA improved 1H and 3He MRI segmentation over the non-DA model (P = 0.007 to 0.03) but GAN-DA did not yield further improvement. The cascaded U-Net improved non-ventilated lung segmentation (P < 0.005). Dice similarity coefficients (DSC) between manually and DL-segmented full lung, non-ventilated, hypo-ventilated, and normal-ventilated regions were 0.965 ± 0.010, 0.840 ± 0.057, 0.715 ± 0.175, and 0.883 ± 0.060, respectively. We observed no statistically significant difference in DCSs between participants with and without COPD (P = 0.41, 0.06, and 0.18 for non-ventilated, hypo-ventilated, and normal-ventilated regions, respectively). CONCLUSION: The proposed cascaded U-Net framework generated fully-automated segmentation of ventilation defects on 3He MRI among older smokers with and without COPD that is consistent with our reference method.