The rapid development of machine learning and artificial intelligence technologies has promoted the widespread application of data-driven algorithms in the field of building energy consumption prediction. This study comprehensively explores diversified prediction strategies for different time scales, building types, and energy consumption forms, constructing a framework for artificial intelligence technologies in this field. With the prediction process as the core, it deeply analyzes the four key aspects of data acquisition, feature selection, model construction, and evaluation. The review covers three data acquisition methods, considers seven key factors affecting building loads, and introduces four efficient feature extraction techniques. Meanwhile, it conducts an in-depth analysis of mainstream prediction models, clarifying their unique advantages and applicable scenarios when dealing with complex energy consumption data. By systematically combing the existing research, this paper evaluates the advantages, disadvantages, and applicability of each method and provides insights into future development trends, offering clear research directions and guidance for researchers.

Background: To assess cancer-specific mortality (CSM) and other-cause mortality (OCM) rates in patients with rare histological prostate cancer subtypes. Methods: Using the Surveillance, Epidemiology, and End Results database (2004–2020), we applied smoothed cumulative incidence plots and competing risks regression (CRR) models. Results: Of 827,549 patients, 1510 (0.18%) harbored ductal, 952 (0.12%) neuroendocrine, 462 (0.06%) mucinous, and 95 (0.01%) signet ring cell carcinoma. In the localized stage, five-year CSM vs. OCM rates ranged from 2 vs. 10% in acinar and 3 vs. 8% in mucinous, to 55 vs. 19% in neuroendocrine carcinoma patients. In the locally advanced stage, five-year CSM vs. OCM rates ranged from 5 vs. 6% in acinar, to 14 vs. 16% in ductal, and to 71 vs. 15% in neuroendocrine carcinoma patients. In the metastatic stage, five-year CSM vs. OCM rates ranged from 49 vs. 15% in signet ring cell and 56 vs. 16% in mucinous, to 63 vs. 9% in ductal and 85 vs. 12% in neuroendocrine carcinoma. In multivariable CRR, localized neuroendocrine (HR 3.09), locally advanced neuroendocrine (HR 9.66), locally advanced ductal (HR 2.26), and finally metastatic neuroendocrine carcinoma patients (HR 3.57; all p < 0.001) exhibited higher CSM rates relative to acinar adenocarcinoma patients. Conclusions: Compared to acinar adenocarcinoma, patients with neuroendocrine carcinoma of all stages and locally advanced ductal carcinoma exhibit higher CSM rates. Conversely, CSM rates of mucinous and signet ring cell adenocarcinoma do not differ from those of acinar adenocarcinoma.