A machine learning based prediction system for the Indian Ocean Dipole.

The Indian Ocean Dipole (IOD) is a mode of climate variability observed in the Indian Ocean sea surface temperature anomalies with one pole off Sumatra and the other pole near East Africa. An IOD event starts sometime in May-June, peaks in September-October and ends in November. Through atmospheric teleconnections, it affects the climate of many parts of the world, especially that of East Africa, Australia, India, Japan, and Europe. Owing to its large impacts, previous studies have addressed the predictability of the IOD using state of the art coupled climate models. Here, for the first-time, we predict the IOD using machine learning techniques, in particular artificial neural networks (ANNs). The IOD forecasts are generated for May to November from February-April conditions. The attributes for the ANNs are derived from sea surface temperature, 850 hPa and 200 hPa geopotential height anomalies, using a correlation analysis for the period 1949-2018. An ensemble of ANN forecasts is generated using 500 samples with replacement using jackknife approach. The ensemble mean of the IOD forecasts indicates the machine learning based ANN models to be capable of forecasting the IOD index well in advance with excellent skills. The forecast skills are much superior to the skills obtained from the persistence forecasts that one would guess from the observed data. The ANN models also perform far better than the models of the North American Multi-Model Ensemble (NMME) with higher correlation coefficients and lower root mean square errors (RMSE) for all the target months of May-November.

Improving seasonal forecasts of air temperature using a genetic algorithm.

Seasonal forecasts of air-temperature generated by numerical models provide guidance to the planners and to the society as a whole. However, generating accurate seasonal forecasts is challenging mainly due to the stochastic nature of the atmospheric internal variability. Therefore, an array of ensemble members is often used to capture the prediction signals. With large spread in the prediction plumes, it becomes important to employ techniques to reduce the effects of unrealistic members. One such technique is to create a weighted average of the ensemble members of seasonal forecasts. In this study, we applied a machine learning technique, viz. a genetic algorithm, to derive optimum weights for the 24-ensemble members of the coupled general circulation model; the Scale Interaction Experiment-Frontier research center for global change version 2 (SINTEX-F2) boreal summer forecasts. Our analysis showed the technique to have significantly improved the 2m-air temperature anomalies over several regions of South America, North America, Australia and Russia compared to the unweighted ensemble mean. The spatial distribution of air temperature anomalies is improved by the GA technique leading to better representation of anomalies in the predictions. Hence, machine learning techniques could help in improving the regional air temperature forecasts over the mid- and high-latitude regions where the model skills are relatively modest.

A stochastic dynamical systems view of the Atlantic Multidecadal Oscillation.

We provide a dynamical systems framework to understand the Atlantic Multidecadal Oscillation and show that this framework is in many ways similar to that of the El Niño/Southern Oscillation. A so-called minimal primitive equation model is used to represent the Atlantic Ocean circulation. Within this minimal model, we identify a normal mode of multidecadal variability that can destabilize the background climate state through a Hopf bifurcation. Next, we argue that noise is setting the amplitude of the sea surface temperature variability associated with this normal mode. The results provide support that a stochastic Hopf bifurcation is involved in the multidecadal variability as observed in the North Atlantic.

Light-induced mass turnover in a mono-species community of mixotrophs.

We formulate a simple model for growth of a facultative photoautotroph with chemoheterotrophic capabilities. The organism is described by zero, one or three reserve components, and one structural component, all taken to be generalized compounds. The rules of synthesizing units are used for interactions among the uptake processes of the various nutrients and light (parallel processing), and for the merging of autotrophic and heterotrophic activities (sequential processing). For simplicity, we focus on the assimilation of inorganic carbon, inorganic nitrogen and light, and of two organic compounds (dead reserves and dead structure) that originate from aging. The process of resource recycling in a closed environment, as driven by light, and its links with community's structure (amount of biomass) is analysed in this simplest of all communities. Explicit analytical expressions for the steady states show how structure and function depend on the system parameters light, total carbon and total nitrogen. The behaviour resembles the Monod model for the Canonical Community, a three-species ecosystem consisting of producers, consumers and decomposers. If trophic preferences of a mixotroph are allowed to follow a random walk across generations, a trophic structure evolves where mixotrophs coexist with auto- and heterotrophs. Depth profiles are presented for the implied steady-state concentrations of dissolved inorganic carbon and nitrogen.

[Breast complaints during the Dutch nationwide breast cancer screening program: increased risk of referral and of breast cancer]. / Borstklachten ten tijde van bevolkingsonderzoek naar borstkanker: grotere kans op verwijzing en op borstkanker.

OBJECTIVE: To determine the prevalence of breast complaints reported on a questionnaire among participants in the first round of the national breast cancer screening programme in the Netherlands, and to assess the relationship between these complaints and referral and diagnosis of breast cancer. DESIGN: Retrospective controlled study. METHOD: Women in the South screening region of the Dutch nation-wide breast cancer screening programme (Stichting Bevolkingsonderzoek Borstkander Zuid (BoBZ)) received a medical questionnaire together with the invitation for participation in the screening programme. This questionnaire was only reviewed by the radiologist if there were doubts concerning referral. The study assessed whether participants with these complaints had higher risks of referral and screen-detected breast cancer than those without these complaints. In this study two random and independent samples were drawn from the screening files: 1116 women who attended the first screening round and 1113 women who were referred for medical checkup in the same screening round. RESULTS: The prevalence of breast complaints reported in the first screening round was 5.1% (95% CI = 3.9-6.6). More than half of the complaints included pain not related to menses; nipple secretion; changes in skin (retraction or discolouration) or tumours. The referral risk for women with these complaints was more than twice that for women without complaints (adjusted odds ratio (OR) = 2.3; 95% CI = 1.4-3.8) and their risk of screen-detected breast cancer was more than three times as high (adjusted OR = 3.3; 95% CI = 1.4-3.8). Up to 5.8% of the screen-detected breast cancers could be attributed to the breast complaints. CONCLUSION: Breast complaints at the time of breast cancer screening increased the risk of breast cancer being detected, but it is not clear as yet whether the screening will benefit from the explicit incorporation of questions regarding complaints.

Influence of magnetic field strength on T2* decay and phase effects in gradient echo MRI of vertebral bone marrow.

OBJECTIVE: The goal of this study was twofold: first, to quantify the influence of magnetic field strength (B0) on T2* and T2susceptibility relaxation rates in vertebral bone marrow; and second, to determine the significance of in-phase/out-of-phase phenomena in GE MRI as a function of TEs at different B0 values. MATERIALS AND METHODS: Magnetic resonance imaging of the spine was performed at 1.5, 1.0, and 0.2 T using a GE sequence (2D-FLASH) with variable TE. Vertebral body signal intensities were measured and plotted against TE values. These curves were used to calculate T2* decay of bone marrow at different B0 values. RESULTS: At all field strengths, we observed exponential signal decay, modulated by a cosine function, the periodicity of which was proportional to B0. The T2* values of vertebral bodies were 9.7 +/- 1.8 ms at 1.5 T, 17.2 +/- 2.4 ms at 1.0 T, and 53 +/- 5 ms at 0.2 T. True T2 relaxation time (independent of field strength) was 62 +/- 4 ms. The T2susceptibility was 11.7 ms at 1.5 T, 24.8 ms at 1.0 T, and 585 ms at 0.2 T. CONCLUSION: We conclude that at low B0, susceptibility effects become less important and T2* approaches true T2 relaxation.

Low-field versus high-field MR imaging of the knee: a comparison of signal behaviour and diagnostic performance.

A prospective study was undertaken to compare MR imaging of the knee obtained with low-field and high-field systems. In 10 subjects, MR imaging of the knee was performed on a 0.2 T permanent magnet and on a 1.5 T superconductive system. Similar spin echo (SE) and 3D-FISP (3D Fourier transform with steady state precession) acquisitions were obtained. Comparative image analysis was performed independently by four radiologists. Results show that the image quality and diagnostic performance delivered by state-of-the-art 0.2 T and 1.5 T systems is equivalent. Advantages of the 1.5 T system included: better signal-to-noise ratio, shorter scan times, better visualization of asymptomatic grade 1 meniscal degeneration on SE images. Advantages of 0.2 T images were: decreased chemical shift, susceptibility and flow artifacts, improved evaluation of subchondral bone on 3D-FISP images, slightly better patient tolerance. We conclude that, for MR imaging of the knee, a low-field system is a cost-effective alternative to more expensive superconducting units.

Graphical method to determine sigmoid ligand binding in vivo at low doses.

In this study, a graphical method is presented to detect sigmoidal binding of in vivo data at low nonsaturating doses. The method can also be applied when the overall binding includes nonspecific binding. Irrespective of the mathematical expression describing the saturation curve, it is shown that a sigmoid curve, in contrast to a nonsigmoid, can be converted into a peak-shaped curve by normalizing the ordinate values with the corresponding value of the abscissa. Computer simulations of in vivo ligand binding demonstrate this principle for in vivo data plots. Experimental results are described to illustrate this approach based on in vivo binding of the neuroleptic drug N-methylspiperone in the rat brain. It is concluded that the method is in particular useful for in vivo applications, e.g., positron emission tomography (PET) studies in human, because low doses are required, and specifically bound, nonspecifically bound, and unbound ligand do not need to be determined separately.

Retrosomatic cleft associated with an invisible lumbar pedicle.

The congenital nature of an invisible lumbar pedicle can easily be recognized by several associated characteristic features. This prevents confusion with other pedicular abnormalities or alterations caused by osteolytic metastases or erosion due to soft tissue tumours. A few variants of typical features are described.

CT manifestations of intracranial involvement by Hodgkin's disease.

CT brain scanning, in a patient diagnosed as having stage IV B nodular sclerosis Hodgkin's disease and since 1978 in complete remission, demonstrated intracranial relapse. Neurological symptoms could be caused by scattered intracranial lesions from Hodgkin's disease. Different CT scan manifestations are described.