SCU-Net: A deep learning method for segmentation and quantification of breast arterial calcifications on mammograms.

PURPOSE: Measurements of breast arterial calcifications (BAC) can offer a personalized, non-invasive approach to risk-stratify women for cardiovascular diseases such as heart attack and stroke. We aim to detect and segment breast arterial calcifications in mammograms accurately and suggest novel measurements to quantify detected BAC for future clinical applications. METHODS: To separate BAC in mammograms, we propose a lightweight fine vessel segmentation method Simple Context U-Net (SCU-Net). Due to the large image size of mammograms, we adopt a patch-based way to train SCU-Net and obtain the final whole-image-size results by stitching patchwise results together. To further quantify calcifications, we test five quantitative metrics to inspect the progression of BAC for subjects: sum of mask probability metric ( P M ), sum of mask area metric ( A M ), sum of mask intensity metric ( S I M ), sum of mask area with threshold intensity metric T A M X , and sum of mask intensity with threshold X metric T S I M X . Finally, we demonstrate the ability of the metrics to longitudinally measure calcifications in a group of 26 subjects and evaluate our quantification metrics compared with calcified voxels and calcium mass on breast CT for 10 subjects. RESULTS: Our segmentation results are compared with state-of-the-art network architectures based on recall, precision, accuracy, F1 score/Dice score, and Jaccard index evaluation metrics and achieve corresponding values of 0.789, 0.708, 0.997, 0.729, and 0.581 for whole-image-size results. The quantification results all show >95% correlation between quantification measures on predicted masks of SCU-Net as compared to the groundtruth and measurement of calcification on breast CT. For the calcification quantification measurement, our calcification volume (voxels) results yield R2 -correlation values of 0.834, 0.843, 0.832, 0.798, and 0.800 for the P M , A M , S I M , T A M 100 , T S I M 100 metrics, respectively; our calcium mass results yield comparable R2 -correlation values of 0.866, 0.873, 0.840, 0.774, and 0.798 for the same metrics. CONCLUSIONS: Simple Context U-Net is a simple method to accurately segment arterial calcification retrospectively on routine mammograms. Quantification of the calcifications based on this segmentation in the retrospective cohort study has sufficient sensitivity to detect the normal progression over time and should be useful for future research and clinical applications.

Simultaneous cardiac and respiratory inhibition during seizure precedes death in the DBA/1 audiogenic mouse model of SUDEP.

This study was designed to evaluate cardiac and respiratory dysfunction in a mouse model of sudden unexpected death in epilepsy i.e., SUDEP. We simultaneously monitored respiration via plethysmography and the electrocardiogram via telemetry before, during, and after an audiogenic seizure. DBA/1 mice responded to an acoustic stimulus with one or two cycles of circling and jumping before entering a clonic/tonic seizure. This was followed by death unless the mice were resuscitated by mechanical ventilation using room air. During the initial clonic phase, respiration declined and cardiac rhythm is slowed. By the tonic phase, respiration had ceased, atrial P-waves were absent or dissociated from the QRS complex, and heart rate had decreased from 771±11 to 252±16 bpm. Heart rate further deteriorated terminating in asystole unless the mice were resuscitated at the end of the tonic phase which resulted in abrupt recovery of P-waves and a return to normal sinus rhythm, associated with gasping. Interestingly, P-waves were preserved in the mice treated with methylatropine during the pre-ictal period (to block parasympathetic stimulation) and heart rate remained unchanged through the end of the tonic phase (765±8 vs. 748±21 bpm), but as in control, methylatropine treated mice died from respiratory arrest. These results demonstrate that a clonic/tonic seizure in the DBA/1 mouse results in abrupt and simultaneous respiratory and cardiac depression. Although death clearly results from respiratory arrest, our results suggest that seizure activates two central nervous system pathways in this model-one inhibits respiratory drive, whereas the other inhibits cardiac function via vagal efferents. The abrupt and simultaneous recovery of both respiration and cardiac function with mechanical ventilation within an early post-ictal timeframe shows that the vagal discharge can be rapidly terminated. Understanding the central mechanism associated with the abrupt cardiorespiratory dysfunction and equally abrupt recovery may provide clues for therapeutic targets for SUDEP.

Alfuzosin delays cardiac repolarization by a novel mechanism.

The United States Food and Drug Administration (FDA) uses alfuzosin as an example of a drug having QT risk in humans that was not detected in nonclinical studies. FDA approval required a thorough clinical QT study (TCQS) that was weakly positive at high doses. The FDA has used the clinical/nonclinical discordance as a basis for mandatory TCQS, and this requirement has serious consequences for drug development. For this reason, we re-examined whether nonclinical signals of QT risk for alfuzosin were truly absent. Alfuzosin significantly prolonged action potential duration (APD)(60) in rabbit Purkinje fibers (p < 0.05) and QT in isolated rabbit hearts (p < 0.05) at the clinically relevant concentration of 300 nM. In man, the QT interval corrected with Fridericia's formula increased 7.7 ms, which exceeds the 5.0-ms threshold for a positive TCQS. Effects on hK(v)11.1, hK(v)4.3, and hK(v)7.1/hKCNE1 potassium currents and calcium current were not involved. At 300 nM, approximately 30x C(max), alfuzosin significantly increased whole-cell peak sodium (hNa(v)1.5) current (p < 0.05), increased the probability of late hNa(v)1.5 single-channel openings, and significantly shortened the slow time constant for recovery from inactivation. Alfuzosin also increased hNa(v)1.5 burst duration and number of openings per burst between 2- and 3-fold. Alfuzosin is a rare example of a non-antiarrhythmic drug that delays cardiac repolarization not by blocking hK(v)11.1 potassium current, but by increasing sodium current. Nonclinical studies clearly show that alfuzosin increases plateau potential and prolongs APD and QT, consistent with QT prolongation in man. The results challenge the FDA grounds for the absolute primacy of TCQS based on the claim of a false-negative, nonclinical study on alfuzosin.

Deferoxamine prevents cardiac hypertrophy and failure in the gerbil model of iron-induced cardiomyopathy.

To evaluate the effects of the iron chelator deferoxamine on the functional and structural manifestations of iron-induced cardiac dysfunction, we measured cardiac power, left ventricular systolic, and diastolic function as (dP/dt)max and (dP/dt)min, respectively, and left ventricular and septal wall thickness in isolated heart preparations derived from the Mongolian gerbil model of iron overload. We induced iron overload with weekly subcutaneous injections of iron dextran (800 mg/kg/wk); deferoxamine (DFO; 100 mg/kg) was administered twice daily by subcutaneous injection, 5 of 7 days each week; and control animals received weekly subcutaneous injections of dextran alone. Animals administered iron alone initially exhibited, at 5 weeks, increased cardiac power but by 12 to 20 weeks, cardiac power was severely diminished, with impairment of both systolic and diastolic function of the left ventricle and marked cardiac hypertrophy (P<.001 for all vs control animals). Administration of DFO with iron did not interfere with the initial augmentation of cardiac power at 5 weeks but prevented the subsequent deterioration in cardiac performance. After 12 to 20 weeks, gerbils given DFO with iron had mean values of cardiac power indistinguishable from those of control animals; both systolic and diastolic function were significantly enhanced not only in comparison with those of animals treated with iron alone but also with respect to controls. In addition, DFO prevented cardiac hypertrophy; mean ventricular and septal wall thickness in gerbils given DFO and iron were not significantly different from those in controls. In the gerbil model of iron overload, concurrent administration of DFO with iron prevents both the development of cardiac hypertrophy and the progressive deterioration in cardiac performance that are produced by chronic iron accumulation.

Optical mapping reveals conduction slowing and impulse block in iron-overload cardiomyopathy.

Cardiac disease with arrhythmia or heart failure is the leading cause of death in patients with thalassemia major and a major complication of other forms of iron overload. Current antiarrhythmic treatment does not appear to alter the clinical course. Using a gerbil model of iron-overload cardiomyopathy, we previously observed a reduction in the fast inward sodium current in isolated cardiomyocytes. Electrocardiograms (ECGs) in the same gerbil model indicate PR-interval prolongation, QRS-interval widening, and arrhythmias. We hypothesize that such changes in the ECG in this model are the result of abnormal action-potential conduction at the level of the whole heart. To test this hypothesis, we took ECGs and recorded action potentials using high-resolution optical mapping from the anterior surface of 9 iron-overloaded and 9 age-matched control ventricular-paced, Langendorff-perfused gerbil hearts. The iron-overloaded gerbils received weekly iron-dextran injections of 800 mg/kg for 14 to 18 weeks. ECGs showed QRS- and PR-interval prolongation in iron-treated gerbils compared with that in controls. In addition, atrioventricular block was observed in 2 of 6 iron-treated gerbils but not in controls. Conduction velocity was significantly slower in iron-treated gerbils than in controls. At normal pacing rates, abnormal activation patterns caused by stable regions of conduction block were observed in iron-overloaded gerbils (33%) but not in controls. Such abnormal impulse conduction may be a mechanism of increased arrhythmia vulnerability in iron-overload cardiomyopathy.

Deferoxamine promotes survival and prevents electrocardiographic abnormalities in the gerbil model of iron-overload cardiomyopathy.

We investigated the time course of electrocardiographic (ECG) changes in the Mongolian gerbil model of iron overload and the effects of the iron chelator deferoxamine (DFO) on these changes. Iron overload was produced with weekly subcutaneous injections of low doses (200 mg/kg/wk) or high doses (800 mg/kg/wk) of iron-dextran. DFO was administered subcutaneously at a dose of 200 mg/kg/day to high-dose animals. Our results show that (1) survival of iron-overloaded gerbils is dose-dependent, with median survival times of 68 and 14 weeks for low- and high-dose animals, respectively; (2) both low and high doses produce prolongation of the PR interval and bradycardia in early stages and prolongation of the QT interval, premature ventricular contractions, variable degrees of atrioventricular block, changes in the ST segment, and T-wave inversion at later stages coinciding with the development of heart failure; (3) DFO prevented death during 20 weeks of high-dose iron-dextran; (4) DFO prevented ECG changes, although delayed prolongation of PR intervals and QRS complexes occurred; and (5) despite marked prolongation of survival and prevention of ECG changes, DFO had modest effects on total cardiac iron content. We speculate that DFO chelates a small iron pool located within the cytoplasm of iron-overloaded cardiomyocytes.

Bimodal cardiac dysfunction in an animal model of iron overload.

Iron-overload cardiomyopathy is the most common cause of death in patients with thalassemia major, yet the associated changes in cardiac function have not been quantified. We studied the effects of iron overload on cardiac function in Mongolian gerbils, a species that responds to iron overload in the same manner as human beings. We injected iron-dextran or dextran alone at low subcutaneous doses (200 mg/kg/wk) for 20 to 60 weeks and at high doses (800 mg/kg/wk) for 6 to 20 weeks. At shorter durations for either dose, the mean values of cardiac work, coronary flow, left ventricular (dP/dt)(max) and left ventricular (dP/dt)(min) in isolated perfused hearts were significantly greater than control values; at longer durations, these values were significantly less than control values. Echocardiography in intact animals showed eccentric cardiac hypertrophy, increased cardiac output, and normal exercise tolerance at shorter durations of dosage. At longer durations, concentric cardiac hypertrophy developed, and cardiac output and exercise capacity were impaired. The response to iron overload in Mongolian gerbils progresses from an initial state of high cardiac output to a subsequent state of low-output failure similar to the course of cardiomyopathy that has been inferred in patients with transfusional iron overload.