Pulmonary congestion is a key determinant of heart failure, but for a long time it has been an elusive target for the clinical cardiologist in the pre-B-line era, despite research efforts of Carlo Giuntini, a pneumologist who attempted the quantification of lung water in the seventies with too insensitive chest X-ray lung water score, too cumbersome nuclear medicine, and too complex invasive thermodilution techniques. Daniel Lichtenstein, is a French intensivist who first discovered lung ultrasound as a sign of extravascular lung water in 1997. B-lines (also known as ultrasound lung comets) detectable by lung ultrasound arise from the pleural line, extend towards the edge of the screen, and move synchronously with respiration. In cardiology, B-lines were introduced in 2004 and are now the dominant technique for research applications and clinical purposes. B-lines showed a prognostic value in several clinical scenarios, largely independent and additive over echocardiographic predictors such as ejection fraction. The methodology became user-friendly in the last years, with a reduction of the scanning sites from the original 28 to a simplified 4-site scan now extracting information on lung water in < 1 minute. More recently, B-lines were also studied during physical and pharmacological stress. Signs of pulmonary congestion are found during stress in 1 out of 3 all-comers with normal findings at rest. Artificial intelligence applied to ultrasound and clinical data allows for the detection of B lines, their quantification, and the assessment of their nature. The B-lines phenotype can cluster around different endotypes: dry (in systemic sclerosis and lung interstitial fibrosis); wet (water); sterile (as in cardiogenic edema); infective (as in COVID-19 and interstitial pneumonia); right heart-sided (as in pulmonary arterial hypertension); left-heart sided (as in heart failure or valvular heart disease). Artificial intelligence B-lines and pocket-size insonation of the B-lines-driven decongestion therapy are now on the horizon.

Ischemic heart disease (IHD) is a major global health issue, frequently resulting in myocardial infarction and ischemic cardiomyopathy. Prompt and precise diagnosis is essential to avert complications such as heart failure and sudden cardiac death. Although invasive coronary angiography remains the gold standard for high-risk patients, noninvasive multimodality imaging is becoming more prevalent for those at low-to-intermediate risk. This review evaluated the current state of multimodality imaging in IHD, including echocardiography, nuclear cardiology, cardiac magnetic resonance imaging (MRI), computed tomography (CT) angiography, and invasive coronary angiography. Each modality has distinct strengths and limitations, and their complementary use provides a comprehensive assessment of cardiac health. Integrating artificial intelligence (AI) into imaging workflows holds promise for enhancing diagnostic accuracy and efficiency. AI algorithms can optimize image acquisition, processing, and interpretation of complex imaging data. Emerging technologies like 4D flow MRI, molecular imaging, and hybrid systems [e.g., positron emission tomography (PET)/MRI, PET/CT] integrate anatomical, functional, and molecular data, providing comprehensive insights into cardiac pathology and potentially revolutionizing the management of IHD. This review also explored the clinical applications and impact of multimodality imaging on patient outcomes, emphasizing its role in improving diagnostic precision and guiding therapeutic decisions. Future directions include AI-driven decision support systems and personalized medicine approaches. Addressing regulatory and ethical challenges, such as data privacy and algorithm transparency, is crucial for the broader adoption of these advanced technologies. This review highlighted the transformative potential of AI-enhanced multimodality imaging in improving the diagnosis and management of IHD.