Search Results (1,973)

Background and Objectives: Rotational atherectomy (RA) is a crucial method for percutaneous coronary intervention (PCI) of heavily calcified coronary lesions. The aim of this study was to compare the clinical outcomes in patients undergoing RA via the radial versus femoral approach. Methods: The Rotational Atherectomy in Calcified Lesions in Korea (ROCK) registry included consecutive patients with severely calcified coronary artery disease who received RA during PCI at nine tertiary centers in Korea. A total of 540 patients who underwent PCI with RA were enrolled between October 2019 and January 2010. We retrospectively investigated the clinical outcomes between the transradial and transfemoral approaches. The primary endpoint was major adverse cardiac and cerebrovascular events (MACCE) within 36 months of follow-up. Results: Of the 540 patients, 248 patients (45.9%) were in the transradial group, and 292 patients (54.1%) were in the transfemoral group. There were no significant differences in MACCE (11.3% vs. 17.8%, adjusted hazard ratio [HR]: 1.520; 95% confidence interval: 0.889–2.600; p = 0.126) and procedural success (97.6% vs. 95.2%, p = 0.145). The occurrence of in-hospital bleeding was numerically higher in the transfemoral group, but the difference was not statistically significant (8 [3.2%] vs. 19 [6.5%], p = 0.081) Conclusions: In this study, the transradial approach did not show a significant difference in clinical outcomes but tended to have lower bleeding events compared to the transfemoral approach. RA via the transradial approach can be a useful vascular access option compared to the transfemoral approach. Full article

Background/Objectives: Virtual reality (VR), a component of extended reality (XR), has shown promise in pre-procedural planning by providing immersive, patient-specific simulations. In pain management, where precise anatomical understanding is critical for interventions such as peripheral nerve stimulation (PNS), nerve blocks, and intrathecal pump placement, the application of VR remains underexplored. This case series examines the role of VR in enhancing pre-procedural planning for complex chronic pain interventions. Methods: From August 2022 to December 2024, six patients with anatomically challenging conditions underwent VR-assisted pre-procedural planning at Weill Cornell Medical Center. Patient-specific 3D models were created using the manual or automatic segmentation of imaging data and reviewed in VR to optimize procedural strategies by the surgeons performing the case. Procedures were then performed using conventional fluoroscopic or ultrasound guidance. Results: In all cases, VR facilitated the improved visualization of complex anatomies and informed optimal procedural trajectories. In patients with a complex cancer anatomy, previous surgical changes, or hardware, VR enabled precise PNS lead or needle placement, resulting in significant pain reductions postoperatively. In certain cases where previous interventional pain procedures had failed, VR allowed for a “second opinion” to develop an alternative approach with improved outcomes. Finally, in one case, VR served to potentially prevent patient harm by providing insight to the proceduralists regarding an alternative approach. Across the series, VR enhanced the spatial awareness, procedural accuracy, and confidence in navigating challenging anatomical scenarios. Conclusions: This case series demonstrates the utility of VR in pre-procedural planning for chronic pain interventions. By enabling detailed anatomical visualization and trajectory optimization, VR has the potential to improve outcomes in complex cases. Further studies are needed to evaluate its broader clinical applications and cost-effectiveness in pain management. Full article

Advances in artificial intelligence (AI), multi-omic profiling, and sophisticated imaging technologies have significantly advanced personalized medicine in gastrointestinal surgical oncology. These technological innovations enable precise patient stratification, tailored surgical strategies, and individualized therapeutic approaches, thereby significantly enhancing clinical outcomes. Despite remarkable progress, challenges persist, including the standardization and integration of diverse data types, ethical concerns regarding patient privacy, and rigorous clinical validation of predictive models. Addressing these challenges requires establishing international standards for data interoperability, such as Fast Healthcare Interoperability Resources, and adopting advanced security methods, such as homomorphic encryption, to facilitate secure multi-institutional data sharing. Moreover, ensuring model transparency and explainability through techniques such as explainable AI is critical for fostering trust among clinicians and patients. The successful integration of these advanced technologies necessitates strong multidisciplinary collaboration among surgeons, radiologists, geneticists, pathologists, and oncologists. Ultimately, the continued development and effective implementation of these personalized medical strategies complemented by human expertise promise a transformative shift toward patient-centered care, improving long-term outcomes for patients with gastrointestinal cancer. Full article

Background: The “fast track” addition (within 48 h) of proprotein convertase subtilisin/kexin type 9 inhibitors (PCSK9i) to the optimized oral lipid-lowering therapy (LLT) during hospitalization for acute coronary syndrome (ACS) has been shown to rapidly achieve the low-density lipoprotein cholesterol (LDL-C) therapeutic targets. However, so far, its efficacy in real-world settings remains understudied. Methods: We retrospectively analyzed 128 ACS patients treated at our center, comparing “PCSK9i fast track” use within 48 h to standard “stepwise” LLT. Lipid levels and incidence of major adverse cardiovascular events (MACEs) were evaluated at 30 and 180 days. Results: The “PCSK9i fast track” group achieved significantly lower LDL-C levels at 30 days (41.5 ± 27.5 vs. 85.6 ± 35.9 mg/dL, p < 0.001) and 180 days (29.6 ± 21.0 vs. 59.0 ± 32.4 mg/dL, p < 0.001). Recommended LDL-C targets (<55 mg/dL) were met by 88.3% of the “PCSK9i fast track” group at 180 days, compared with 61.9% of controls (p < 0.001). No significant differences in MACEs were observed between groups. No adverse effects from PCSK9i use were noted. Conclusions: The “PCSK9i fast track” strategy was safe and effective in achieving LDL-C targets more rapidly than conventional approaches in real-world ACS patients. Full article

Introduction: Tuberous sclerosis complex (TSC) is a rare multisystemic genetic disorder characterized by the formation of benign tumors in various organs, including the central nervous system, skin, kidneys, and heart. The diagnosis is based on well-defined clinical criteria, such as those from Schwartz (2007) updated in 2012 by the International Tuberous Sclerosis Complex Consensus Group. The study aims to investigate the clinical, imaging, and molecular characteristics of patients diagnosed with tuberous sclerosis and to explore the correlation between specific genetic mutations (TSC1 and TSC2 genes) and the severity of clinical manifestations. Material and Methods: This is a retrospective longitudinal study of 13 patients diagnosed with tuberous sclerosis, identified in the records of the Bihor Regional Center for Medical Genetics (BRCMG) within the Bihor County Emergency Clinical Hospital from 1984 to 2024. Clinical, imaging, and molecular features were assessed. Patients were evaluated by a multidisciplinary team, including a geneticist, pediatrician, neurologist, psychiatrist, and psychologist. Clinical and imaging data were retrospectively collected from the congenital malformations and genetic disease records of BRCMG Bihor and statistically analyzed. Results: All patients showed clinical and imaging signs consistent with the diagnosis of tuberous sclerosis. Neurological manifestations were present in 83% of patients, including epilepsy and cognitive delays. Renal lesions were detected in 46% of cases, and dermatological lesions, such as facial angiofibromas, were observed in 69% of patients. Mutational variants identified in the TSC2 gene correlated with a more severe clinical presentation, including severe intellectual disability and treatment-resistant seizures, compared to variants in the TSC1 gene. Conclusions: Our study, although involving a small number of patients, highlights the clinical heterogeneity of tuberous sclerosis and the importance of a multidisciplinary approach in patient management. Early diagnosis and ongoing monitoring are essential to improving the quality of life for patients. Further studies are needed to assess the impact of therapeutic interventions and genetic correlations within the studied population. Full article

Background: Atrial flutter (AFL) is a macro-reentrant tachycardia classified as cavotricuspid isthmus (CTI)-dependent or non-CTI-dependent based on its reliance on the CTI for conduction. CTI dependence can present as type I ECG (sawtooth flutter waves in inferior leads and positive P-waves in V1) or type II ECG (absence of these characteristics). This study aimed to identify clinical and electrophysiological parameters to improve CTI dependence prediction in AFL. Methods: Patients at the Ulm University Heart Center between 2010 and 2019 with AFL undergoing electrophysiological studies and ablation were enrolled. Clinical and electrophysiological parameters such as age, gender, prior comorbidities, interventions, and medication use were analyzed. Results: The study included 383 patients, with 70% presenting with type I ECG AFL. CTI dependence was observed in 242 (90.3%) type I ECG patients and 52 (45.2%) type II ECG patients. CTI-dependent AFL patients were younger and had fewer comorbidities. Predictors for CTI dependence in type I ECG included male gender (p = 0.006), absence of beta-blocker use (p = 0.031), no prior atrial fibrillation (p = 0.035), and no prior pulmonary vein isolation (p < 0.001). In type II ECG, predictors for CTI dependence included younger age (p = 0.016), male gender (p = 0.007), absence of arterial hypertension (p = 0.036), and longer atrial cycle length (p < 0.001). Conclusions: Identifying clinical and electrophysiological parameters enhances the ability to predict CTI dependence in AFL, offering valuable insights for tailored diagnostic and therapeutic approaches. Coupling these parameters with ECG findings holds promise for refining prediction accuracy and optimizing patient care. Full article

Background and Objectives: Distal radius fractures (DRFs) are among the most common fractures globally, with a lifetime incidence of around 9%. They typically present in two age peaks: high-impact trauma in patients under 40 and low-energy trauma in those over 40. Intra-articular DRFs are classified according to the Arbeitsgemeinschaft für Osteosynthesefragen (AO) classification, influencing the treatment approach. Surgical management, particularly open reduction and internal fixation (ORIF) using volar plate osteosynthesis, is considered the gold standard. This study aims to compare the treatment costs of fluoroscopy-assisted ORIF and arthroscopy-assisted ORIF for intra-articular DRF. The analysis includes surgical procedure costs, material expenses, and operating time to evaluate the cost-effectiveness of both methods, considering reimbursement within the German healthcare system. Materials and Methods: A retrospective, monocentric study was conducted at Ludwig-Maximilians-University (LMU) Hospital, a supraregional hand trauma center in southern Germany. Patients with DRFs requiring ORIF were treated either with fluoroscopy or arthroscopic assistance. Group 1 included patients treated by the Department of Hand Surgery (Plastic Surgery), subdivided into Group 1a (arthroscopy-assisted) and Group 1b (fluoroscopy-only). Group 2 comprised patients treated by Orthopaedics and Trauma Surgery (fluoroscopy-only). Costs associated with surgical procedures, including materials, operating time, and postoperative care, were analyzed. Results: A total of 43 DRFs were treated. Group 1 consisted of 17 cases, with an average age of 49.6 years (SD = 19.4) and a 64% majority of female patients. Of these, 10 cases were treated with arthroscopy-assisted ORIF (Group 1a) and 7 with fluoroscopy-only ORIF (Group 1b). In Group 1a, the average age was 53.9 years (SD = 16.3) with 60% female and 40% male patients, while in Group 1b, the average age was 43.6 years (SD = 23.1) with 71.4% female patients. Group 2 included 25 cases, with an average age of 54.2 years (SD = 21.0) and a distribution of 64% female and 36% male patients. There was no significant difference in age and gender distribution within the groups and subgroups (p > 0.05). The mean procedure time was longer for arthroscopically assisted ORIF (111.5 min) compared to fluoroscopy-only ORIF (80.1 min), and even longer compared to Group 2 (65.0 min). Material costs were slightly higher in Group 1. Total costs for Group 1 averaged EUR 4906.58, with subgroup costs of EUR 5448.24 for arthroscopy-assisted and EUR 4132.80 for fluoroscopy-only. In comparison, Group 2 costs averaged EUR 3344.08. Conclusions: Intra-articular DRFs with severely displaced fragments or concomitant injuries benefit from arthroscopically assisted fracture treatment. While material costs do not significantly differ between arthroscopically assisted and fluoroscopy-only treatments, the significantly longer procedure time for arthroscopy-assisted ORIF results in the largest cost component. Despite this, reimbursement through the DRG system remains fixed and does not account for the increased operative duration or complexity of arthroscopic procedures. Our findings demonstrate that DRF treatment, regardless of the method used, is either not or only marginally cost-covering under the current German reimbursement structure. In the context of the ongoing shift towards outpatient hand surgery, including the management of DRF, adequate reimbursement rates are necessary to ensure the economic viability of DRF management, particularly for complex intra-articular fractures requiring arthroscopic assistance. Full article

Background/Objectives: An anastomotic leak (AL) following colorectal surgery is one of the most feared complications due to its associated morbidity and mortality. Early detection of ALs remains difficult, as the development of clinical signs of deterioration can be a late finding. This is particularly problematic in patients with poor access to care after discharge. C-reactive protein (CRP) is a systemic marker of inflammation that has been proposed as an early AL screening. However, absolute cut-off values have been shown to have limited sensitivity and specificity. We propose the use of CRP trends for early AL detection. Methods: A retrospective chart review of patients undergoing surgery requiring at least one anastomosis at a single tertiary care center was performed. Patients with two or fewer postoperative CRP values were excluded. Postoperative CRP trends were compared between control and AL patients using a mixed model with a Geisser–Greenhouse correction. Results: CRP trends differed significantly between AL and control patients, with a 10% CRP increase after postoperative day two showing 100% sensitivity and 84% specificity for an AL as well as a 100% negative predictive value. Accepted CRP cut-off values on postoperative days three and four had sensitivities of only 71.4% and 80% and specificities of 70.0% and 76.5%, respectively. CRP trends differed in AL versus control patients despite the surgical approach or presence of additional procedures. Conclusions: Daily monitoring of CRP trends (versus absolute cut-offs) may enhance early anastomotic leak detection and aid in discharge decision-making, particularly important in rural settings with limited healthcare access. Full article

Background/Objectives: MVI is a relevant prognostic factor among patients with hepatocellular carcinoma (HCC) receiving liver transplantation (LT). The preoperative assessment of the risk for MVI is relevant to pre-LT patient management and selection. The objective of this study was to create and validate a model to predict microvascular invasion (MVI) based on preoperative variables in the LT setting. Methods: A total of 2170 patients from 11 collaborative centers in Europe, Asia, and the US, who received transplants between 1 January 2000 and 31 December 2017, were enrolled in the study. The entire cohort was split into a training and a validation set (70/30% of the initial cohort, respectively) using random selection. Results: MVI was reported in 586 (27.0%) explanted specimens. Using the training set data, multivariable logistic regression identified three preoperative parameters associated with MVI: α-fetoprotein (lnAFP; odds ratio [OR] = 1.19; 95% confidence interval [CI] = 1.13–1.27), imaging tumor burden score (lnTBS; OR = 1.66; 95%CI = 1.39–1.99), and a fast-track approach before LT due to the availability of a live donation (OR = 1.99; 95%CI = 1.56–2.53). In the validation set, the LT-MVI c-index was 0.74, versus 0.69 for the MVI score proposed by Endo et al. (Brier Skill Score +75%). The new score had a relevant net reclassification index (overall value = 0.61). Stratifying the validation set into three risk categories (0–50th, 51st–75th, and >75th score percentiles), a very good stratification was observed in terms of disease-free (5-year: 89.3, 75.5, and 50.7%, respectively) and overall survival (5-year: 79.5, 72.6, and 53.7%, respectively). Conclusions: The preoperative assessment of MVI using the proposed score demonstrated very good accuracy in predicting MVI after LT. Full article

Background: Comorbidities such as hypertension and hypercholesterolemia are risk factors associated with Mild Cognitive Impairment (MCI) and Alzheimer’s disease (AD). The most significant genetic risk factor is the ε4 allele of the apolipoprotein E gene (APOE). The aim of this paper is to determine whether hypertension is the most significant but modifiable risk factor to delay AD onset. Method: A cohort of patients with MCI (N = 3052) is developed from the documented database (N = 43,999) within the National Alzheimer’s Coordinating Center (NACC) during the time period from June 2005 to May 2021. Cox proportional hazard models with propensity score weights on demographic information and comorbidities at baseline are applied to examine association of hypertension and hypercholesterolemia with AD onset among MCI patients. Associations are compared to APOE genotypes and AD onset. In addition, the association of hypertension with decline rates in Mini-Mental State Examination (MMSE) scores are reported. Results: After controlling for age, sex, race, APOEε4, and reported comorbidities, the results show that MCI patients who subsequently develop hypertension within 18 months after their first diagnosis of MCI have a significantly higher risk of AD onset (HR = 2.77, 95%CI (1.66, 4.65), p value < 0.0001), compared to MCI patients with no hypertension or a late occurrence of hypertension after 18 months. This significant association is validated through a Random Forest method, a machine learning approach with bootstrap simulations. In addition, patients with early hypertension have significantly higher MMSE score declining rates compared to those without hypertension (coefficient = 0.988, p = 0.0054.). Conclusions: Hypertension is the most significant risk factor comparable to the genetic risk factor APOEε4 allele. Our finding is unique, as we did not observe a similar outcome in those with early hypercholesterolemia. Thus, among all comorbidities, hypertension is the most significant risk factor similar to the genetic risk factor APOEε4 allele. Full article

(1) Background: Mixed Reality (MR) technology, such as the HoloLens, offers a novel approach to preoperative education. This study evaluates its feasibility and effectiveness in improving patient comprehension and comfort during informed consent for pancreatoduodenectomy. (2) Methods: A single-center, randomized, controlled pilot study was conducted between February and May 2023. Patients recommended for pancreatoduodenectomy were randomized into a control group receiving standard education or an intervention group using the HoloLens. Pre- and post-intervention surveys assessed patient understanding and comfort. (3) Results: Nineteen patients participated (8 HoloLens, 11 control). Both groups showed improved comprehension post-intervention, but only the HoloLens group demonstrated a statistically significant increase (Z = −2.524, p = 0.012). MR users had a greater understanding of surgical steps compared to controls, and 75% of participants in both groups reported high comfort levels with the surgery. MR integration was feasible and did not disrupt clinical workflow. (4) Conclusions: These findings suggest that MR can enhance preoperative education for complex procedures. However, limitations include the small sample size and single-center design, necessitating larger studies to confirm its broader applicability. MR-based education represents a promising tool to improve patient engagement and comprehension in surgical decision making. Full article

On behalf of the Canadian Association of Psychosocial Oncology, we are pleased to present the Abstracts from the 2025 Annual Conference, titled “Responding to the Human Experience of Cancer and Caring for the Soul: Building on 40 years of global leadership in psychosocial oncology”. The 40th Annual CAPO Conference was held in Toronto from 23 April 2025 to 25 April 2025. In an era marked by the rapid advancement of biologically focused precision medicine, it is imperative to redirect our attention towards the human experience of illness and the soul of medicine. Biomedicine has conceptualized illness in ways that have proved profoundly productive from a curative and biological point of view. But it cannot—and it does not pretend to—illuminate the experience of living with it. (Hurwitz 2009). This conference aims to delve into the intricate interplay between cutting-edge biomedical technologies inclusive of artificial intelligence and big data and the deeply personal narratives of individuals navigating illness. By shifting the focus from mere disease pathology to encompassing the holistic human experience, we aspire to foster a more compassionate and patient-centered approach to healthcare with psychosocial support at the core of humanistic care that can improve survival and well-being in all aspects of a whole-person approach to illness. Through interdisciplinary dialogue and introspection, we endeavor to illuminate the profound connection between mind, body, and spirit in the practice of medicine, reaffirming the timeless significance of empathy, understanding, and human connection in healing and psychosocial aspects of care as fundamental to living well with cancer. This conference brought together key stakeholders including multidisciplinary professionals from nursing, psychology, psychiatry, social work, spiritual care, nutrition, medicine, rehabilitation medicine, occupational health and radiation therapy for both adult and pediatric populations. Participants included clinicians, researchers, educators in cancer care, community-based organizations and patient representatives. Patients, caregivers and family members presented abstracts that speak to their role in managing cancer experiences and care. Over two hundred (200) abstracts were submitted for presentation as symposia, 20-minute oral presentations, 10-minute oral presentations, 90-minute workshops and poster presentations. We congratulate all the presenters on their research work and contribution. Full article

Background/Objectives: Over the past three decades, significant progress has been made in understanding the biology of neuroblastoma. The integration of prognostic factors has facilitated risk stratification and the development of targeted treatment approaches. This study aims to analyze the outcomes of pediatric patients with neuroblastoma treated at a single oncology center over a 30-year period. Methods: This retrospective study analyzed data from patients aged 0–18 years with neuroblastoma, treated at the Department of Pediatric Hematology and Oncology in Bydgoszcz, Poland, between 1993 and 2023. The study endpoints included the 5-year probability of overall survival (pOS), event-free survival (pEFS), and relapse-free survival (pRFS), analyzed separately for low/intermediate- and high-risk groups. Results: Seventy-five patients met the inclusion criteria. Thirty-two children were categorized as high-risk patients and forty-three as low/intermediate risk. During the study period, outcomes in the low/intermediate-risk group improved significantly (the 5-year pOS 85.7% vs. 100.0%, p = 0.019; the 5-year pRFS 85.7% vs. 100.0%, p = 0.662; the 5-year pEFS 83.3% vs. 100.0%, p = 0.038). In the high-risk group, outcomes improved but did not reach statistical significance (the 5-year pOS 0.0% vs. 41.1%, p = 0.342; the 5-year pRFS 0.0% vs. 32.5%, p = 0.180; and the 5-year pEFS 0.0% vs. 21.5%, p = 0.537). Sixteen patients experienced relapse, of whom only three survived; stem cell transplantation at relapse significantly improved survival (OS 0.0% vs. 50.0%, p = 0.001). In the multivariable analysis, stage at diagnosis was a prognostic factor for pOS (HR 6.0; 95%CI 0.7–49.6, p = 0.096), while pelvic localization was a risk factor for pRFS (HR 3.0; 95%CI 0.8–10.5; p = 0.084). Conclusions: This analysis highlights significant advancements in the diagnosis and treatment of neuroblastoma. Nevertheless, outcomes for high-risk patients and those who experience relapse remain poor, underscoring the need for further therapeutic improvements. Full article

Background/Objectives: Few large cohorts with relatively uniform treatment approaches and long-term follow-up are available for assessing clinical outcomes for breast cancer (BC) patients. The Institut Català d’Oncologia (ICO) Breast Cancer Cohort was designed to well characterize treatment patterns and overall survival outcomes at 5 and 10 years, with a particular focus on patients < 40 and ≥70 years old, age groups often underrepresented in clinical trials. Methods: In this retrospective, observational study, we included all pathologically confirmed invasive BC patients diagnosed and treated between 2010 and 2014 at ICO, a Spanish reference cancer center, with a follow-up until November 2023. We collected comprehensive real-world data on clinicopathologic characteristics and treatment modalities. Overall survival (OS) was estimated using the Kaplan–Meier technique and was reported stratified by prognostic factors for the age groups of ≤40, 41–69 and ≥70. The Multivariate Cox model was used to estimate the risk of death for subgroups of age, adjusting for subtype, stage and grade. Results: Overall, 3451 patients with stage I to IV BC were diagnosed and treated, with a mean age of 58 years (range 19–98); 371 (10.8%) were diagnosed ≤40 years, and 756 (21.9%) were ≥70 years. With a mean follow-up of 9.9 years (SD = 3.5), the 5- and 10-year OS were 89% (95% CI: 86–92%) and 85% (95% CI: 81–88%) for patients ≤ 40, respectively; for those aged 41–69 years, 91% (95% CI: 90–92%) and 85% (95% CI: 83–86%), respectively; and 70% (95% CI: 66–73%) and 50% (95% CI: 47–54%) for those ≥70 years, respectively. The 5- and 10-year relative survival (RS) were 92% and 88% for patients < 70 years, respectively, and 82% and 77% for those ≥70 years, respectively. The Multivariate Cox model identified a HR of 4.90 (95% CI: 3.44–6.97, p < 0.001) for patients ≥ 70 years compared to those between 41 and 69 years. Conclusions: The ICO Breast Cancer Cohort, as far as we know, the largest in Spain with long-term follow-up, underscores the critical role of age and subtype in determining overall survival outcomes in patients with breast cancer. Full article

Background/Objectives: Artificial intelligence (AI) is increasingly influencing oncological research by enabling precision medicine in ovarian cancer through enhanced prediction of therapy response and patient stratification. This systematic review and meta-analysis was conducted to assess the performance of AI-driven models across three key domains: genomics and molecular profiling, radiomics-based imaging analysis, and prediction of immunotherapy response. Methods: Relevant studies were identified through a systematic search across multiple databases (2020–2025), adhering to PRISMA guidelines. Results: Thirteen studies met the inclusion criteria, involving over 10,000 ovarian cancer patients and encompassing diverse AI models such as machine learning classifiers and deep learning architectures. Pooled AUCs indicated strong predictive performance for genomics-based (0.78), radiomics-based (0.88), and immunotherapy-based (0.77) models. Notably, radiogenomics-based AI integrating imaging and molecular data yielded the highest accuracy (AUC = 0.975), highlighting the potential of multi-modal approaches. Heterogeneity and risk of bias were assessed, and evidence certainty was graded. Conclusions: Overall, AI demonstrated promise in predicting therapeutic outcomes in ovarian cancer, with radiomics and integrated radiogenomics emerging as leading strategies. Future efforts should prioritize explainability, prospective multi-center validation, and integration of immune and spatial transcriptomic data to support clinical implementation and individualized treatment strategies. Unlike earlier reviews, this study synthesizes a broader range of AI applications in ovarian cancer and provides pooled performance metrics across diverse models. It examines the methodological soundness of the selected studies and highlights current gaps and opportunities for clinical translation, offering a comprehensive and forward-looking perspective in the field. Full article