Artificial intelligence (AI) is, in essence, the concept of ‘computer thinking’, encompassing methods that train computers to perform and learn from executing certain tasks, called machine learning, and methods to build intricate computer models that both learn and adapt, called complex neural networks. Computer vision is a function of AI by which machine learning and complex neural networks can be applied to enable computers to capture, analyze, and interpret information from clinical images and visual inputs. This annotation summarizes key considerations and future perspectives concerning computer vision, questioning the need for this technology (the ‘why’), the current applications (the ‘what’), and the approach to unlocking its full potential (the ‘how’).

Cite this article: Bone Joint J 2022;104-B(8):911–914.

Aims

The number of convolutional neural networks (CNN) available for fracture detection and classification is rapidly increasing. External validation of a CNN on a temporally separate (separated by time) or geographically separate (separated by location) dataset is crucial to assess generalizability of the CNN before application to clinical practice in other institutions. We aimed to answer the following questions: are current CNNs for fracture recognition externally valid?; which methods are applied for external validation (EV)?; and, what are reported performances of the EV sets compared to the internal validation (IV) sets of these CNNs?

Aims

Recently, several synovial biomarkers have been introduced into the algorithm for the diagnosis of a prosthetic joint infection (PJI). Alpha defensin is a promising biomarker, with a high sensitivity and specificity, but it is expensive. Calprotectin is a protein that is present in the cytoplasm of neutrophils, is released upon neutrophil activation and exhibits anti-microbial activity. Our aim, in this study, was to determine the diagnostic potential of synovial calprotectin in the diagnosis of a PJI.

Vascularised fibular grafts (VFGs ) are a valuable surgical technique in limb salvage after resection of a tumour. The primary objective of this multicentre study was to assess the risk factors for failure and complications for using a VFG after resection of a tumour.

The study involved 74 consecutive patients (45 men and 29 women with mean age of 23 years (1 to 64) from four tertiary centres for orthopaedic oncology who underwent reconstruction using a VFG after resection of a tumour between 1996 and 2011. There were 52 primary and 22 secondary reconstructions. The mean follow-up was 77 months (10 to 195).

In all, 69 patients (93%) had successful limb salvage; all of these united and 65 (88%) showed hypertrophy of the graft. The mean time to union differed between those involving the upper (28 weeks; 12 to 96) and lower limbs (44 weeks; 12 to 250). Fracture occurred in 11 (15%), and nonunion in 14 (19%) patients.

In 35 patients (47%) at least one complication arose, with a greater proportion in lower limb reconstructions, non-bridging osteosynthesis, and in children. These complications resulted in revision surgery in 26 patients (35%).

VFG is a successful and durable technique for reconstruction of a defect in bone after resection of a tumour, but is accompanied by a significant risk of complications, that often require revision surgery. Union was not markedly influenced by the need for chemo- or radiotherapy, but should not be expected during chemotherapy. Therefore, restricted weight-bearing within this period is advocated.

Cite this article: Bone Joint J 2015;97-B:853–61.

Atypical cartilaginous tumours are usually treated by curettage. The purpose of this study was to show that radiofrequency ablation was an effective alternative treatment.

We enrolled 20 patients (two male, 18 female, mean age 56 years (36 to 72) in a proof-of-principle study. After inclusion, biopsy and radiofrequency ablation were performed, followed three months later by curettage and adjuvant phenolisation. The primary endpoint was the proportional necrosis in the retrieved material. Secondary endpoints were correlation with the findings on gadolinium enhanced MRI, functional outcome and complications.

Our results show that 95% to 100% necrosis was obtained in 14 of the 20 patients. MRI had a 91% sensitivity and 67% specificity for detecting residual tumour after curettage. The mean functional outcome (MSTS) score six weeks after radiofrequency ablation was 27.1 (23 to 30) compared with 18.1 (12 to 25) after curettage (p < 0.001). No complications occurred after ablation, while two patients developed a pathological fracture after curettage.

We have shown that radiofrequency ablation is capable of completely eradicating cartilaginous tumour cells in selective cases. MRI has a 91% sensitivity for detecting any residual tumour. Radiofrequency ablation can be performed on an outpatient basis allowing a rapid return to normal activities. If it can be made more effective, it has the potential to provide better local control, while improving functional outcome.

Cite this article: Bone Joint J 2014;96-B:1540–5.

In this case study, we describe the clinical presentation and treatment of 36 patients with periosteal chondrosarcoma collected over a 59-year period by the archive of the Netherlands Committee on Bone Tumours. The demographics, clinical presentation, radiological features, treatment and follow-up are presented with the size, location, the histological grading of the tumour and the survival.

We found a slight predominance of men (61%), and a predilection for the distal femur (33%) and proximal humerus (33%). The metaphysis was the most common site (47%) and the most common presentation was with pain (44%). Half the tumours were classified histologically as grade 1. Pulmonary metastases were reported in one patient after an intra-lesional resection. A second patient died from local recurrence and possible pulmonary and skin metastases after an incomplete resection.

It is clearly important to make the diagnosis appropriately because an incomplete resection may result in local recurrence and metastatic spread. Staging for metastatic disease is recommended in grade II or III lesions.

These patients should be managed with a contrast-enhanced MRI of the tumour and histological confirmation by biopsy, followed by en-bloc excision.

Cite this article: Bone Joint J 2014;96-B:823–8.

There has been an increase in the incidence of bone and joint tuberculosis (BJTB) in The Netherlands and we have carried out an epidemiological study in order to find an explanation for this increase. Data from 1993 to 2000 from The Netherlands Tuberculosis Register (NTR) were used. In 1993 there was a total of 52 patients with BJTB. This figure increased gradually to 80 in 1999 before decreasing to 61 in 2000.

There was a total of 12 447 patients with tuberculosis; BJTB was found in 532, accounting for 4.3% of all cases and 10.6% of all extrapulmonary cases. Localisation in the spine occurred in 56%.

Certain immigrants, in particular from Somalia, were more likely to have BJTB than other immigrants or the native Dutch population. Increased age and female gender were associated with BJTB. Only 15% of BJTB patients also suffered from pulmonary tuberculosis. The usual long delay in the diagnosis of BJTB may be shortened if physicians are more aware of tuberculosis.