Evaluation of risk factors and survival rates of patients with early-stage breast cancer with machine learning and traditional methods.

BACKGROUND: This article is aimed to make predictions in terms of prognostic factors and compare prediction methods by using Cox proportional hazards regression analysis (CPH), some machine learning techniques and Accelerated Failure Time (AFT) model for post-treatment survival probabilities according to clinical presentations and pathological information of early-stage breast cancer patients. MATERIAL AND METHODS: The study was carried out in three stages. In the first stage, the CPH method was applied. In the second stage, the AFT model and in the last stage, machine learning methods were applied. The data set consists of 697 breast cancer patients who applied to Marmara University Hospital oncology clinic between 01.01.1994 and 31.12.2009. The models obtained by using various parameters of the patients were compared according to the C index, 5-year survival rate and 10-year survival rate. RESULTS AND CONCLUSION: According to the models obtained as a result of the analyses applied, MetLN and age were obtained as a significant risk factor as a result of CPH method and AFT methods, while MetLN, age, tumor size, LV1 and extracapsular involvement were obtained as risk factors in machine learning methods. In addition, when the c-index values of the handheld models are examined, it is obtained as 69.8 for the CPH model, 70.36 for the AFT model, 72.1 for the random survival forest and 72.8 for the gradient boosting machine. In conclusion, the study highlights the potential of comparing conventional statistical methods and machine-learning algorithms to improve the precision of risk factor determination in early-stage breast cancer prognosis. Additionally, efforts should be made to enhance the interpretability of machine-learning models, ensuring that the results obtained can be effectively communicated and utilized by clinical practitioners. This would enable more informed decision-making and personalized care in the treatment and follow-up processes for early-stage breast cancer patients.

The John Hopkins classification system used in pediatric supracondylar humerus fractures requires more experience than the Gartland system.

The Gartland classification is used to decide on surgery. In contrast, the John Hopkins classification system predicts clinical outcomes for patients undergoing surgery and determines the risk of reduction loss. This study aims to investigate the usability of the Gartland and the John Hopkins classification systems by pediatric and general orthopedic surgeons. The preoperative images of 200 patients who presented at a tertiary-level trauma center with a supracondylar humerus fracture were examined by 4 observers, twice at an interval of 6 weeks. The observers comprised 2 pediatric orthopedic surgeons and 2 orthopedics and traumatology surgeons. Inter- and intra-observer reliability were excellent for the first and second measurements of the Gartland classification. The inter-observer agreement of the pediatric orthopedic surgeons and the general orthopedic surgeons was found to be similar (ICC >90). Inter-observer reliability was excellent (ICC: 0.808) and good (ICC: 0.732) for the measurements of the John Hopkins classification, respectively. The inter-observer agreement of the pediatric orthopedic surgeons and the general orthopedic surgeons was not similar. The inter-observer agreement between the two pediatric orthopedic surgeons was excellent for the measurements (ICC: 0.868; ICC: 0.756, respectively). The inter-observer agreement between the two general orthopedic surgeons was good for the measurements (ICC: 0.605; ICC: 0.663, respectively). The John Hopkins classification system has good intra- and inter-observer reliability, and a high experience level increases the agreement. The Gartland classification system was not affected by experience. This should be considered when taking measurements in studies and patient management.

Factors influencing study outcomes in recent literature on distal radial fracture treatment.

The main tools used by an orthopaedic surgeon for managing distal radial fracture treatment are recent literature and treatment guidelines. The aim of the present study was to find which factors within the study design influence study outcomes the most. Trials in three major databases (PubMed, Scopus, Embase) comparing surgical and non-surgical treatment options for adolescent and adult distal radial fractures with their original data, between 2013 and 2021, were included. The selected 47 studies were classified according to their outcomes. The relationship between study characteristics and outcomes was statistically analysed. It was more likely to find no difference in outcomes between volar locking plate and less invasive treatments when the sample size was above 100, follow-up was more than 1 year and functional assessments were used. A small sample size and short follow-up time affect study outcomes in favour of a volar locking plate. Readers should focus on the design criteria and read the full text of the studies before making any conclusions. LEVEL OF EVIDENCE: III.

Tibial Tuberosity Rotation in Patients With Patellar Instability Versus Age- and Sex-Matched Controls.

Background: Several methods have been described to measure external rotation of the tibial tuberosity; all use femoral landmarks. Purpose: To develop reproducible tibial-based methods to measure external rotation of the tibial tuberosity in patients with patellar instability. Study Design: Cross-sectional study; Level of evidence, 3. Methods: Included were magnetic resonance images of 61 patients with patellar instability and 61 age- and sex-matched healthy controls. Three novel methods using tibial landmarks to measure the rotation of the tibial tuberosity (plateau axis-tuberosity axis [PA-TA] angle, tibial geometric center-tuberosity axis [GC-TA] angle, and plateau axis-midtuberosity point [PA-MT] angle) as well as a femoral-based rotational measurement (surgical transepicondylar axis-tuberosity axis [sTEA-TA] angle) and the tibial tuberosity-trochlear groove (TT-TG) distance were measured and compared in instability patients and controls using unpaired t tests, and the cutoff values for predicting instability were calculated using receiver operating characteristic curves. The correlations between the angle measurements and the TT-TG distance were analyzed. Results: Instability patients had significantly higher external rotation of the tibial tuberosity compared with controls with regard to the PA-TA angle (18.2° ± 9.6° versus 13.1° ± 6.8°; P = .001), GC-TA angle (8.4° ± 4.5° versus 11.5° ± 3.9°; P = .0001) and sTEA-TA angle (122° ± 8.5° versus 113.6° ± 6.3°; P = .0001). The mean TT-TG distance was also significantly higher in the instability group (18.2 ± 5.4 versus 11.5 ± 2.7 mm; P = .001). The cutoff values were 17.5° (area under the receiver operating characteristic curve [AUC] = 0.66) for PA-TA angle, 8.5° (AUC = 0.705) for GC-TA angle, 118.8° (AUC = 0.79) for sTEA-TA angle, and 15.2 mm for TT-TG distance (AUC = 0.863). PA-TA angle was significantly correlated with all other measurements (r = 0.35-0.71; P ≤ .006 for all), whereas sTEA-TA angle had the strongest correlation with TT-TG distance (r = 0.78; P = .001). Conclusion: The tibial tuberosity was externally rotated in patellar instability patients compared with age- and sex-matched controls, and this intrinsic malalignment of the proximal tibia was demonstrated in the tibial-based measurements.

Reliability of Gordon Lateral Rotation Percentage and Prabhakar Percentage of Metaphyseal Overhang for Pediatric Supracondylar Humerus Fracture; Is it Clinically Reliable for John Hopkins Classification Coronal Fracture Subtypes?

BACKGROUND: Rotational malalignment is a common problem in pediatric supracondylar humerus fractures (SCHF). Several techniques have been described to evaluate the true rotation value. Although the Prabhakar and Gordon techniques are used frequently, their superiority to each other in terms of measurement quality is unknown. QUESTIONS/PURPOSES: The aim of this study was to investigate the clinical compatibility of the 2 techniques and to evaluate whether they are suitable for all subtypes. METHODS: This cross-sectional study included 40 patients with SCHF (including subtypes; Typical, Medial Oblique, Lateral Oblique, and High fracture pattern). The Gordon lateral rotation percentage and Prabhakar percentage of metaphyseal overhang were measured twice by 4 experienced Orthopedics and Traumatology surgeons at 8-week intervals. The interobserver and intraobserver reliability were examined using the intraclass correlation coefficient. RESULTS: The interobserver reliability for Gordon and Prabhakar technique was 0.816 and 0.762 for the first measurement and 0.811 and 0.811 for the second measurement, respectively.The medial oblique fracture pattern was determined to have the best interobserver agreement among the subtypes. The result was excellent for the medial and lateral oblique subtypes, good for the typical fracture pattern, and fair for the high fracture pattern. The intraobserver reliability for Gordon and Prabhakar technique was excellent, 0.924 and 0.922, respectively. CONCLUSION: The main finding of this study was that the Gordon and Prabhakar techniques have similar interobserver and intraobserver reliability. Although the Gordon technique tends to have higher interobserver reliability, the difference was clinically insignificant. These measurements should not be relied upon in cases of SCHF with a high fracture pattern because of the different anatomic features of that region. LEVEL OF EVIDENCE: Level III.

Foot and Ankle Forklift Injuries; Diagnosis to Treatment Options, Return to Work and Functional Outcomes.

As our tertiary-level trauma center is close to an industrial zone and associated international port, industrial injuries are seen more often than is expected in routine practice. The aim of this study was to present the demographic data, treatment options, and changes in work-life of patients with foot and ankle injuries due to a forklift accident, and to determine the factors affecting the clinical outcome. The study was conducted in our university hospital and included patients who presented with a foot and ankle injury resulting from a forklift accident. Data were collected including age, sex, dominant extremity, history of trauma, presence of fracture, management (surgical or conservative), complications, time of return to work, workforce loss, and the American Orthopaedic Foot and Ankle Society (AOFAS) and Short Form Health Survey (SF-36) scores at the final follow-up examination. Evaluation was made of 132 patients, comprising 113 (85.6%) males and 19 (14.4%) females with a mean age of 32 years (range 16-65 years). The most frequently recorded occupational group was shipyard workers (50%). The most common mechanism of injury was crushing under the forklift wheel (n = 63, 47.7%). The most frequently injured foot region was the forefoot (47%). A total of 90 (68.1%) patients continued to work in the same position on return to work. The lowest mean AOFAS score (73.4) and SF-36 (physical component) score (37.3) were determined in cases with mixed region injuries (p = .0001, p = .0001). The wearing of protective footwear had no effect on the rate of return to work (p = .195), workforce loss (p = .34) and AOFAS score (p = .166). This study is the largest series of patients with foot and ankle injuries related to forklift accidents. Forklift injuries can be treated conservatively or surgically according to the clinical condition of the patient. The main indicators of return to work and functional outcome are which foot region is injured and whether or not the injury causes a fracture.