Thymectomy plays an important role in patients with myasthenia gravis (MG). This study aimed to explore predictors of postoperative myasthenic crisis (POMC) after thymectomy and to define a predictive score of respiratory failure. The clinical data of 177 patients with MG undergoing thymectomy from January 1995 to December 2011 were retrospectively reviewed. The following factors were analysed in relation to the occurrence of myasthenic crisis: gender, age, body mass index (BMI), anti-acetylcholine receptor-antibody level, bulbar symptoms, comorbidities, duration of symptoms, Osserman-stage, Myasthenia Gravis Foundation of America (MGFA) stage, history of myasthenic crisis, use of immoglobulins or plasmapheresis, kind of therapy, spirometric and blood gas parameters, histology, kind of surgery, non-myasthenic complications and duration of intubation. Twenty-two patients experienced postoperative respiratory failure after thymectomy. Univariate analysis revealed a correlation with age >60 years (odds ratio (OR) = 1.79, 95% confidence interval (CI) = 1.04–6.78; P = 0.040); Osserman-stage (IIB- OR = 5.16, 95% CI = 1.10–24.18; P = 0.037, III–IV- OR = 8.75, 95% CI = 1.53–50.05; P = 0.015); bulbar symptoms (OR = 7.42, 95% CI = 1.67–32.84; P = 0.008); BMI >28 (OR = 3.99, 95% CI = 1.58–10.03; P = 0.003); preoperative plasmapheresis (OR = 2.97, 95% CI = 1.18–14.04; P = 0.021); duration of symptoms >2 years (OR = 4.00, 95% CI = 1.09–14.762; P = 0.036); extended surgery (OR = 2.52, 95% CI = 1.02–6.22; P = 0.045); lung (OR = 4.05, 95% CI = 1.44–11.42; P = 0.008), pericardial (OR = 3.78, 95% CI = 1.45–9.82; P = 0.006) or pleural resection (OR = 3.23, 95% CI = 1.30–8.03; P = 0.012); Vital Capacity % 40 mmHg (OR = 3.76, 95% CI = 1.12–12.68; P = 0.032). Multivariate logistic regression analysis showed that Osserman-stage (IIB- OR = 5.69, 95% CI = 1.09–29.69; P = 0.039 (III–IV- OR = 11.33, 95% CI = 1.67–76.72; P = 0.013), BMI >28 (OR = 3.65, 95% CI = 1.10–12.15; P = 0.035), history of myasthenic crisis (OR = 24.10, 95% CI = 2.34–248.04; P = 0.007), duration of symptoms >2 years (OR = 5.94, 95% CI = 1.12–31.48; P = 0.036) and lung resection (OR = 8.48, 95% CI = 2.18–32.97; P = 0.002) independently predict POMC. Excluding history of preoperative myasthenic crisis (statistically associated with Osserman-stage), we built a scoring system according to the OR of Osserman-stage (I–IIA, IIB, III–IV), BMI (<28, ≥28), duration of symptoms (2 years) and association with a pulmonary resection. This model helped in creating four classes with increasing risk of respiratory failure (Group I, 6%; Group II, 10%; Group III, 25%; Group IV, 50%). Our model facilitates the stratification of patient risk and prediction of the occurrence of POMC. Moreover, it could help to guide the anaesthesiologist's decision on the duration of intubation. Further studies based on larger series are needed to confirm these preliminary data.

With profound effects on patient care, the role of artificial intelligence (AI) in radiomics has become a disruptive force in contemporary medicine. Radiomics, the quantitative feature extraction and analysis from medical images, offers useful imaging biomarkers that can reveal important information about the nature of diseases, how well patients respond to treatment and patient outcomes. The use of AI techniques in radiomics, such as machine learning and deep learning, has made it possible to create sophisticated computer-aided diagnostic systems, predictive models, and decision support tools. The many uses of AI in radiomics are examined in this review, encompassing its involvement of quantitative feature extraction from medical images, the machine learning, deep learning and computer-aided diagnostic (CAD) systems approaches in radiomics, and the effect of radiomics and AI on improving workflow automation and efficiency, optimize clinical trials and patient stratification. This review also covers the predictive modeling improvement by machine learning in radiomics, the multimodal integration and enhanced deep learning architectures, and the regulatory and clinical adoption considerations for radiomics-based CAD. Particular emphasis is given to the enormous potential for enhancing diagnosis precision, treatment personalization, and overall patient outcomes.