Breast cancer is a significant health issue affecting women worldwide, and accurately detecting lymph node metastasis is critical in determining treatment and prognosis. While traditional diagnostic methods have limitations and complications, artificial intelligence (AI) techniques such as machine learning (ML) and deep learning (DL) offer promising solutions for improving and supplementing diagnostic procedures. Current research has explored state-of-the-art DL models for breast cancer lymph node classification from radiological images, achieving high performances (AUC: 0.71-0.99). AI models trained on clinicopathological features also show promise in predicting metastasis status (AUC: 0.74-0.77), whereas multimodal (radiomics + clinicopathological features) models combine the best from both approaches and also achieve good results (AUC: 0.82-0.94). Once properly validated, such models could greatly improve cancer care, especially in areas with limited medical resources. This comprehensive review aims to compile knowledge about state-of-the-art AI models used for breast cancer lymph node metastasis detection, discusses proper validation techniques and potential pitfalls and limitations, and presents future directions and best practices to achieve high usability in real-world clinical settings.

UNASSIGNED: Neoadjuvant treatment leads in a reduction in positive lymph nodes and examined lymph nodes (ELN), which may affect assessment of lymph node staging and postoperative treatment. We aimed to compare the staging systems of lymph node ratio (LNR), the positive logarithm ratio of lymph nodes (LODDS), negative lymph nodes (NLN), and the 8th AJCC ypN stage for patients with gastric adenocarcinoma after neoadjuvant therapy.
UNASSIGNED: Data was collected from the Surveillance, Epidemiology, and End Results database and 1,551 patients with gastric adenocarcinoma who underwent neoadjuvant therapy and radical surgery were enrolled. Harrell\'s concordance index, the Receiver Operative Curve, the likelihood ratio test, and the Akaike information criterion were used to compare the predictive abilities of the different staging systems.
UNASSIGNED: Among the 1,551 patients, 689 (44.4%) had ELN < 16 and node-negative patients accounted for 395 (25.5%). When regarded as the categorical variable, LNR had better discrimination power, higher homogeneity, and better model fitness for CSS and OS compared to other stage systems, regardless of the status of ELN. When regarded as the continuos variable, LODDS outperformed others for CSS. Furthermore, the NLN staging system performed superior to others in node-negative patients.
UNASSIGNED: LNR had a better predictive performance than ypN, LODDS and NLN staging systems regardless of the status of ELN when regarded as the categorical variable, whereas LOODS became the better predictive factor for CSS when regarded as the continuos variable. In node-negative patients, NLN might be a feasible option for evaluating prognosis. A combination of LNR and NLN should be considered as user-friendly method in the clinical prognostic assessment.

BACKGROUND: Even though numerous novel lymph node (LN) classification schemes exist, an extensive comparison of their performance in patients with resected pancreatic ductal adenocarcinoma (PDAC) has not yet been performed.
METHODS: We investigated the prognostic performance and discriminative ability of 25 different LN ratio (LNR) and 27 log odds of metastatic LN (LODDS) classifications by means of Cox regression and C-statistic in 319 patients with resected PDAC. Regression models were adjusted for age, sex, T category, grading, localization, presence of metastatic disease, positivity of resection margins, and neoadjuvant therapy.
RESULTS: Both LNR or LODDS as continuous variables were associated with advanced tumor stage, distant metastasis, positive resection margins, and PDAC of the head or corpus. Two distinct LN classifications, one LODDS and one LNR, were found to be superior to the N category in the complete patient collective. However, only the LODDS classification exhibited statistically significant, gradually increasing HRs of their subcategories and at the same time significantly higher discriminative potential in the subgroups of patients with PDAC of the head or corpus and in patients with tumor free resection margins or M0 status, respectively. On this basis, we built a clinically helpful nomogram to estimate the prognosis of patients after radically resected PDAC.
CONCLUSIONS: One LNR and one LODDS classification scheme were found to out-perform the N category in terms of both prognostic performance and discriminative ability, in distinct patient subgroups, with reference to OS in patients with resected PDAC.

BACKGROUND: Lymph node ratio (LNR) and the Log odds of positive lymph nodes (LODDS) have been proposed as a new prognostic indicator in surgical oncology. Various studies have shown a superior discriminating power of LODDS over LNR and lymph node category (N) in diverse cancer entities, when examined as a continuous variable. However, for each of the classification systems various cut-off values have been defined, with the question of the most appropriate for patients with CRC still remaining open. The present study aimed to compare the predictive impact of different lymph node classification systems and to define the best cut-off values regarding accurate evaluation of overall survival in patients with resectable, non-metastatic colorectal cancer (CRC).
METHODS: CRC patients who underwent surgical resection from 1996 to 2018 were extracted from our medical data base. Cox proportional hazards regression models and C-statistics were performed to assess the discriminative power of 25 LNR and 26 LODDS classifications. Regression models were adjusted for age, sex, extent of the tumor, differentiation, tumor size and localization.
RESULTS: Our study group consisted of 654 consecutive patients with non-metastatic CRC. C-statistic revealed 2 LNR and 5 LODDS classifications that demonstrated superior prognostic performance in patients with UICC III CRC, compared to the N category. No clear advantage of one classification over another could be demonstrated in any other patient subgroup.
CONCLUSIONS: Distinct LNR and LODDS classifications demonstrate a prognostic superiority over the N category only in patients with Stage III radically resected CRC.

In this feasibility study we aimed to evaluate the value of previously reported molecular tumor biomarkers associated with lymph node metastasis in oral squamous cell carcinoma (OSCC) to optimize neck strategy selection criteria.
The association between expression of cortactin, cyclin D1, FADD, RAB25, and S100A9 and sentinel lymph node status was evaluated in a series of 87 (cT1-2N0) patients with OSCC treated with primary resection and SLNB procedure.
Tumor infiltration depth and tumor pattern of invasion were independent prognostic markers for SLN status, while none of the tumor makers showed a better prognostic value to replace SLNB as neck staging technique in the total cohort. However, in the subgroup of patients with pT1N0 OSCC, cortactin expression (OR 16.0, 95%CI 2.0-127.9) was associated with SLN classification.
Expression of cortactin is a promising immunohistochemical tumor marker to identify patients at low risk that may not benefit from SLNB or END.

The aim of this study was to assess the prognoses of patients with non-small cell lung cancer (NSCLC) according to the current nodal (N) categories of the tumor, node and metastasis (TNM) classification and the number of involved lymph node stations.
Five hundred and seventy patients with NSCLC underwent surgery from 1 January 2005 to 31 December 2009 and were analysed retrospectively. Postoperative overall survival was analysed according to two nodal classifications: the current N0, N1, N2 and N3 categories and those based on the number of involved nodal stations: N0, N1a (single N1), N1b (multiple N1), N2a1 (single N2 without N1), N2a2 (single N2 with N1), N2b1 (multiple N2 without N1) and N2b2 (multiple N2 with N1).
Five-year survival rates were 76.1%, 53.4% and 26.3% for N0, N1 and N2, respectively (P < 0.001). When survival was analysed by the number of involved nodal stations, the groups with significant differences were maintained; otherwise, they were merged, and new codes were assigned as follows for exploratory analyses: NA (N0), NB (N1a), NC (N1b, N2a (i.e., N2a1 and N2a2) and N2b1) and ND (N2b2). Five-year survival rates were 76.1%, 60.0%, 39.1%, and 11.4% for NA, NB, NC and ND, respectively, and there were significant differences among them. This N classification was an independent prognostic factor in multivariate analyses.
Pending prospective and international validation, it is practical to merge the current N categories with the number of involved lymph node stations when evaluating the postoperative prognosis of NSCLC patients.

To evaluate the effectiveness, reproducibility, and usability of our proposed nodal nomenclature and classification system employed for several years in our high-volume thyroid cancer unit, for the adequate localization and mapping of lymph nodes in thyroid cancer patients with extensive nodal disease.
Retrospective review.
Thirty-three thyroid cancer patients with extensive nodal disease treated from January 2004 to May 2013 were included in our study. Preoperative ultrasound and computed tomography scans of these patients were reanalyzed by blinded radiologists to investigate the feasibility for the assignment of abnormal lymph nodes to compartments defined in our proposed nodal classification system and to identify areas of difficulty in the assignment.
Analysis of nodal localization revealed a discrepancy in compartment agreement between the two radiologists in the assignment of abnormal nodes in nine patients (9/33, 27%). In six patients (6/33, 18%), discrepancy existed in labeling paratracheal and pretracheal nodes. In three patients (3/33, 9%), disagreement arose in the classification of retrocarotid nodes into lateral versus central compartment. A further refinement of the definition of key borderline regions of the pretracheal versus paratracheal and retrocarotid regions of our classification improved the agreement and demonstrated a complete concordance (100%) amongst the reviewing radiologists.
The proposed nodal classification system, derived specifically for differentiated thyroid carcinoma, with readily identifiable anatomic boundaries on imaging and at surgery, facilitates communication among multidisciplinary physicians and aids in creating a uniform and reproducible radiographic nodal map to guide surgical therapy.
4 Laryngoscope, 127:2429-2436, 2017.