Improved Fracture Recognition with AI assistance

 A retrospective study published in Radiology that included 489 patients with fractures that were interpreted by 24 readers showed a 10% improvement of fracture detection (75% vs 65%, superiority P<.001) when Artificial intelligence (AI) assistance was used.  AI also decreases the reading time by 6.3 seconds.  

The authors concluded that AI assistance improved the sensitivity and specificity of fracture detection for radiologists and non-radiologists alike  and shortened slightly their interpretation time. 

Artificial Intelligence Can Detect Wrist Fractures

A study published in Radiology: Artificial Intelligence found that convolutional neural networks could detect and show fractures on wrist radiographs with a high level of sensitivity and specificity.

A dataset of 7356 wrist radiographs was split into training (90%) and validation (10%) sets.  The models were tested on an unseen test set of 524 consecutive emergency wrist radiographic studies with two radiologists in consensus as the reference standard.

The model detected and correctly localized 310 (91.2%) of 340 and 236 (96.3%) of 245 of all radius and ulna fractures on the frontal and lateral views, respectively. The per-study sensitivity, specificity were 98.1%, and 72.9%, respectively.

The authors concluded that convolutional neural networks were able to detect and localize radius and ulna fractures on wrist radiographs with high sensitivity and specificity.

Ultrasound is ineffective in the healing of bone fractures

A study published in the BMJ suggests that low-intensity pulsed ultrasound, used to speed healing of bone fractures, may be ineffective.

A randomized clinical trial that involved 501 patients who had surgical repair of fractures of the tibia found that patients treated with the low-intensity pulsed ultrasound (LIPUS) healed at the same rate as those given a sham treatment.

Patients self administer daily LIPUS (n=250) or use a sham device (n=251) until their tibial fracture showed radiographic healing or until one year after intramedullary fixation.

Primary registry specified outcome was time to radiographic healing within one year of fixation; secondary outcome was rate of non-union. Additional protocol specified outcomes included short form-36 (SF-36) physical component summary (PCS) scores, return to work, return to household activities, return to ≥80% of function before injury, return to leisure activities, time to full weight bearing, scores on the health utilities index (mark 3), and adverse events related to the device.

Results showed no impact on SF-36 PCS scores between LIPUS and control groups or for the interaction between time and treatment; minimal important difference is 3-5 points or in other functional measures. There was also no difference in time to radiographic healing. There were no differences in safety outcomes between treatment groups. Patient compliance was moderate; 73% of patients administered ≥50% of all recommended treatments.

It was concluded that postoperative use of LIPUS after tibial fracture fixation does not accelerate radiographic healing and fails to improve functional recovery.