Machine Learning Diagnosis of Peritonsillar Abscess

Abstract

Peritonsillar abscess (PTA) is a difficult diagnosis to make clinically, with clinical examination of even otolaryngologists showing poor sensitivity and specificity. Machine learning is a form of artificial intelligence that “learns” from data to make predictions. We developed a machine learning classifier to predict the diagnosis of PTA based on patient symptoms. We retrospectively collected clinical data and symptomatology from 916 patients who underwent attempted needle aspiration for PTA. Machine learning classifiers were trained on a subset of the data to predict the presence or absence of purulence on attempted aspiration. The performance of the model was evaluated on a holdout set. The accuracy of the top-performing algorithm, the artificial neural network, was 72.3%. Artificial neural networks can use patient symptoms to exceed human ability to predict PTA in patients with clinical suspicion for PTA. Similar models can assist medical decision making for clinicians who have suspicion of PTA.

Medical Intervention Alone vs Surgical Drainage for Treatment of Peritonsillar Abscess: A Systematic Review and Meta‐analysis

“…Three studies investigated pain differences between the medical and surgical groups. 3,26,28 Nwe and Singh 30 examined trismus at 15 minutes and 24 hours posttreatment, as well as the ability to swallow water at 2 hours and 24 hours posttreatment. Trismus was 5% improved 15 minutes after antibiotic administration, compared to 48% after aspiration and 100% after incision and drainage.…”

Recent Advances in the Application of Artificial Intelligence in Otorhinolaryngology-Head and Neck Surgery

“…AI has also been used to support clinical diagnoses and treatments, decision-making, the Table 4. Continued prediction of prognoses [98][99][100]125,126], disease profiling, the construction of mass spectral databases [43,[127][128][129], the identification or prediction of disease progress [101,105,[107][108][109][110]130], and the confirmation of diagnoses and the utility of treatments [102][103][104]112,131]. Although many algorithms have been applied, some are not consistently reliable, and certain challenges remain.…”

“…In medical device-based analyses, AI is used to evaluate tissue and blood test results, as well as the outcomes of otorhinolaryngology-specific tests (e.g., polysomnography) [ 72 , 73 , 122 ] and audiometry [ 123 , 124 ]. AI has also been used to support clinical diagnoses and treatments, decision-making, the prediction of prognoses [ 98 - 100 , 125 , 126 ], disease profiling, the construction of mass spectral databases [ 43 , 127 - 129 ], the identification or prediction of disease progress [ 101 , 105 , 107 - 110 , 130 ], and the confirmation of diagnoses and the utility of treatments [ 102 - 104 , 112 , 131 ].…”

Machine Learning and Medical Diagnosis.

“…

Recently, Wilson et al (2019) reported the performance of a machine learning classifier to predict the diagnosis of Peritonsillar abscess (PTA) based on patient symptoms. PTA is the most common head and neck abscess; yet, its clinical presentation is imprecise and its management commonly involves needle aspiration or incision and drainage.Clinicians decide whether to aspirate or to drain based on concerning signs and symptoms, such as sore throat, trismus, otalgia, unilateral palatal fullness, and uvular deviation.

…”

“…Recently, Wilson et al (2019) reported the performance of a machine learning classifier to predict the diagnosis of Peritonsillar abscess (PTA) based on patient symptoms. PTA is the most common head and neck abscess; yet, its clinical presentation is imprecise and its management commonly involves needle aspiration or incision and drainage.…”